ArticlePDF Available

It’s Not That You Said It, It’s How You Said It: Exploring the Linguistic Mechanisms Underlying Values Affirmation Interventions at Scale

Authors:

Abstract and Figures

Over the last decade, psychological interventions, such as the values affirmation intervention, have been shown to alleviate the male-female performance difference when delivered in the classroom, however, attempts to scale the intervention are less successful. This study provides unique evidence on this issue by reporting the observed differences between two randomized controlled implementations of the values affirmation intervention: (a) successful in-class and (b) unsuccessful online implementation at scale. Specifically, we use natural language processing to explore the discourse features that characterize successful female students’ values affirmation essays to gain insight on the underlying mechanisms that contribute to the beneficial effects of the intervention. Our results revealed that linguistic dimensions related to aspects of cohesion, affective, cognitive, temporal, and social orientation, independently distinguished between males and females, as well as more and less effective essays. We discuss implications for the pipeline from theory to practice and for psychological interventions.
Content may be subject to copyright.
AERA Open
January-December 2021, Vol. 7, No. 1, pp. 1 –19
DOI:https://doi.org/10.1177/23328584211011611
Article reuse guidelines: sagepub.com/journals-permissions
© The Author(s) 2021. https://journals.sagepub.com/home/ero
Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons
Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use,
reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open
Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
In the American higher education system, achievement gaps
between male and female students persist despite gradual
progress, and are particularly pronounced in STEM fields
(i.e., science, technology, engineering, and mathematics;
Miyake et al., 2010). For instance, across several STEM dis-
ciplines women consistently earn lower exam grades and
lower scores on standardized tests of conceptual mastery
(Brewe et al., 2010; Creech & Sweeder, 2012; Eddy et al.,
2014; Matz et al., 2017; Pollock et al., 2007; Tai & Sadler,
2001). This is counterintuitive given that an important sys-
tematic bias exists in the population of males and females
who attend college, wherein college-bound women system-
atically have higher high school GPAs than men who attend
college (Conger & Long, 2010). This implies that, all else
being equal, females ought to do better in their college
classes than males (Eddy & Brownell, 2016). Despite this,
research has consistently shown gendered performance dif-
ferences (GPDs) that favor males, with male students out-
performing their female counterparts in STEM courses. In
particular, these observed GPDs in education endure even
when accounting for various measures of prior performance,
including high school GPA, standardized tests, and prior col-
lege performance (Eddy & Brownell, 2016; Koester et al.,
2016; Matz et al., 2017). While the causes and consequences
of underachievement of female students in STEM are numer-
ous and complex, the GPD has undoubtedly contributed to
women remaining underrepresented in leadership roles
across all STEM disciplines (National Research Council of
the National Academies, 2011; National Science Foundation,
2019). Social identity threat has consistently been shown to
be one factor which contributes to these GPDs and features
a psychological basis (Dasgupta & Stout, 2014; Steele et al.,
2002).
To address this issue, the current study focuses on evaluat-
ing the effectiveness of the values affirmation (VA) interven-
tion for reducing stereotype threat and improving performance
for female students in STEM. We explicitly focus on gender
and not on other demographic variables (e.g., race, ethnicity,
socioeconomic status) because the STEM courses under
investigation have shown to have significant GPD, wherein
It’s Not That You Said It, It’s How You Said It: Exploring the Linguistic
Mechanisms Underlying Values Affirmation Interventions at Scale
Nia M. M. Dowell
University of California
Timothy A. McKay
University of Michigan
George Perrett
New York University
Over the last decade, psychological interventions, such as the values affirmation intervention, have been shown to alleviate
the male-female performance difference when delivered in the classroom, however, attempts to scale the intervention are less
successful. This study provides unique evidence on this issue by reporting the observed differences between two randomized
controlled implementations of the values affirmation intervention: (a) successful in-class and (b) unsuccessful online imple-
mentation at scale. Specifically, we use natural language processing to explore the discourse features that characterize suc-
cessful female students’ values affirmation essays to gain insight on the underlying mechanisms that contribute to the
beneficial effects of the intervention. Our results revealed that linguistic dimensions related to aspects of cohesion, affective,
cognitive, temporal, and social orientation, independently distinguished between males and females, as well as more and less
effective essays. We discuss implications for the pipeline from theory to practice and for psychological interventions.
Keywords: values affirmation intervention, natural language processing, gender differences in STEM, educational data
science
1011611EROXXX10.1177/23328584211011611Dowell et al.Mechanisms Underlying Values Affirmation Interventions
research-article20212021
Dowell et al.
2
male students consistently outperformed their female coun-
terparts. Toward this effort, we provide a novel assessment of
student-generated VA essays using Educational Data Science
and Learning Analytics techniques. In particular, we capture
the language and discourse properties of students’ VA essays
using two established natural language processing (NLP)
tools, Coh-Metrix (Graesser et al., 2004; McNamara et al.,
2014) and linguistic inquiry word count (LIWC; Pennebaker
et al., 2015; described in a later section). We explore the dif-
ferences in the content of affirmation essays as a function of
gender and successful and not successful VA intervention
implementations. In doing so, we demonstrate how these two
analytical techniques complement each other in the assess-
ment of VA interventions. Although both are established ana-
lytical approaches within the learning analytics community,
thus far, the unique combination of these approaches has not
been utilized in the context of educationally focused psycho-
logical interventions. As such, this study provides unique evi-
dence on this issue by reporting the observed differences
between two randomized field implementations of the VA
intervention at scale: (a) a successful traditional in-class
intervention and (b) an unsuccessful online implementation.
The subsequent sections of the article are organized as
follows. First, we provide a discussion of the psychological
interventions situated within the context of relevant litera-
ture on the VA intervention and the underlying theoretical
framework. Second, we move on to outline the promise of
NLP for psychological interventions situated within the lim-
ited current efforts. We then provide an overview of the cur-
rent research, before moving into the methodological
features of the current investigation, including the principal
component analyses (PCAs) that were used to identify spe-
cific writing profiles and mixed effects analyses to address
the four research questions. Finally, we conclude the article
with a detailed discussion of the results in the context of
theory, as well as a general discussion of the theoretical,
methodological, and practical implications for peer interac-
tion research.
Psychological Interventions at Scale: A Way Forward?
Stereotype threat is a well-established social-psychologi-
cal phenomenon. When an individual is placed in an evalua-
tive environment in which they know others might expect
them to confirm a negative stereotype (e.g., implicit stereo-
types that engineering is a masculine field), they expend
some cognitive resources on this concern, modestly reduc-
ing their ability to perform. Indeed, several studies suggest
that students who feel at risk of upholding stereotypes or
being judged based on stereotypes (i.e., stereotype threat)
experience lower academic performance (Jordt et al., 2017;
Nguyen & Ryan, 2008; Steele & Aronson, 1995).
Struggling with the issues induced by stereotype threat,
either consciously or unconsciously, can prove detrimental for
student performance by reducing working memory (Schmader
& Johns, 2003), which can activate hypervigilance (Forbes
et al., 2008), and consequentially may distract students from
tasks. The issues brought on by stereotype threat can be par-
ticularly detrimental for performance on challenging tasks
(Beilock et al., 2007), such as high-stakes exams that require
more of a student’s mental faculties. The implications of stu-
dents experiencing stereotype threat are not limited to short-
term impacts, such as those on working memory. Indeed,
stereotype threat can have nontrivial, long-term impacts, such
as students distancing themselves from a discipline with
which they once identified (Dasgupta, 2011; Dasgupta et al.,
2015; Fogliati & Bussey, 2013; Thoman et al., 2013; van
Veelen et al., 2019).
This disassociation, coupled with lower performance,
could contribute to a student’s decision to leave STEM and
the resulting underrepresentation of minorities and women
in STEM. For instance, Dasgupta et al. (2015) investigated
the experience of female engineering students in teams of
varying gender ratios: female-minority, sex-parity, and
female-majority and found that female students participated
more actively, and felt less threat/anxiety in female-majority
groups than female-minority groups with sex-parity groups
in-between. Moreover, when assigned to female-minority
groups, women who harbored implicit masculine stereo-
types about engineering reported less confidence and engi-
neering career aspirations.
Implications of Stereotype Threat for Women in STEM
The long-term consequences highlighted by aforemen-
tioned studies and other research (e.g., Cheryan et al., 2009;
Dasgupta, 2011; Dasgupta & Stout, 2014; London et al.,
2011; Thoman & Sansone, 2016; van Veelen et al., 2019)
raises several concerns. A growing proportion of employ-
ment in the United States requires expertise in STEM.
Despite the demand for a STEM-educated workforce, insuf-
ficient numbers of U.S. college graduates have STEM exper-
tise, producing a substantial and persistent gap between
demand and supply (National Science Board, 2016; Skrentny
& Lewis, 2013). This workforce shortage problem is inter-
twined with an equity problem because the undersupply of
Americans who have STEM degrees is larger for women
than for men (Dasgupta & Stout, 2014; National Science
Board, 2015; National Science Foundation, 2019). Fewer
women pursue academic majors and jobs in STEM relative
to their proportions in the U.S. population, even though
these jobs are growing rapidly, lucrative, and of high value.
The relative scarcity of women entering and persisting in
STEM majors in college limits their opportunities to access
high-demand jobs in science, technology, and engineering
after graduation, slowing down socioeconomic mobility.
Clearly, women are untapped human capital that, if lever-
aged, could increase the STEM workforce substantially.
Accomplishing this goal involves identifying academic
stages in the STEM pipeline, where women are less likely to
Mechanisms Underlying Values Affirmation Interventions
3
enter STEM fields and more likely to exit these fields than
men, and developing interventions to address this “leaky
pipeline.” There have been considerable research efforts
devoted to using psychological interventions to address this
problem. Indeed, since the discovery of stereotype threat in
the 1990s, social psychologists have developed a variety of
interventions, which reduce its effects during evaluations.
These include interventions such as the growth mind-set,
utility-value belonging, and VA, which is the focus of the cur-
rent research. Walton (2014) referred to them as “wise inter-
ventions” because they are wise to specific underlying
psychological processes that contribute to social problems or
prevent people from flourishing. Wise interventions are brief,
low-cost interventions that can be implemented in a variety
of contexts and address a psychological need or process that
is responsible for negative outcomes (Casad et al., 2018). In
recent years, an important body of literature has emerged to
explain why such brief interventions may create lasting
impacts (see Harackiewicz & Priniski, 2018, for a review).
However, far less research has explored the underlying
mechanisms of these psychological interventions that result
in more or less beneficial outcomes for women. To address
this issue, the present study focuses on evaluating the effec-
tiveness of the VA intervention for reducing stereotype threat
and improving performance for female students in STEM.
Values Affirmation Intervention
The VA intervention is based on self-affirmation theory
(Steele, 1988), which argues that individuals are motivated to
maintain an overall sense of self-integrity. That is, how indi-
viduals maintain the integrity of the self, especially when it
comes under threat, forms the heart of self-affirmation theory
(Aronson et al., 1999; Sherman & Cohen, 2006; Steele,
1988). Under this perspective, self-affirmations bring about a
more expansive view of the self and an individual’s available
resources (see Cohen & Sherman, 2014, for a review). They
can involve simple everyday activities. In this context, spend-
ing quality time with friends, participating in volunteer activ-
ities, or attending religious services all aid in securing a sense
of adequacy in a higher purpose. According to the self-affir-
mation theory, these affirmations remind individuals of psy-
chosocial resources beyond a specific threat and as such
broaden their perspective beyond it (Sherman & Hartson,
2011). Under normal circumstances, people tend to narrow
their attention on an immediate threat (e.g., the possibility of
not meeting expectations), a response that promotes swift
self-protection. But when self-affirmed, students are able to
reorient and see the many anxieties of daily life in the context
of the big picture (Schmeichel & Vohs, 2009). As such, the
specific threat and the associated implications for the self
become less potent and attract less attention.
The VA intervention has been a widely used strategy to
improve educational outcomes (Casad et al., 2018;
Harackiewicz & Priniski, 2018). Although several versions
of self-affirmation exist, the most examined experimental
manipulation has students write about core personal values
(McQueen & Klein, 2006; Napper et al., 2009), this imple-
mentation was also used in the current research. Personal val-
ues are the internalized standards used to evaluate the self
(Cohen & Sherman, 2014). During the intervention, students
first review a list of values and then choose a few values most
important to them. The list typically excludes values relevant
to a domain of threat (e.g., physics, biology, etc.) in order to
broaden a student’s focus beyond that context. For example,
if a student experiences threats to their identity in an impor-
tant academic domain, such as a woman taking a physics test,
then their self-integrity around this topic is called into ques-
tion. To buffer against threatening negative gender stereo-
types (e.g., women are bad at physics), physics-related
information would be excluded from the list. Students then
write a brief essay about why the selected values are impor-
tant to them and a time when they were important. Thus, a
key aspect of the affirmation intervention is that its content is
self-generated text and tailored to tap into each student’s par-
ticular valued identity (Sherman, 2013). Often students write
about their relationships with friends and family, but they
also frequently write about religion, humor, and kindness. A
central tenet of the VA intervention is that when they affirm
their core values in a threatening environment, students rees-
tablish a perception of personal integrity and worth, which in
turn can provide them with the internal resources needed for
coping effectively (Miyake et al., 2010).
The VA intervention has been shown to have a beneficial
impact on closing achievement gaps in STEM for threatened
groups, such as African American and Hispanic middle and
high school students (Borman et al., 2020; Cohen et al.,
2009; Sherman et al., 2013), undergraduate minority stu-
dents (Brady et al., 2016; Jordt et al., 2017), first-generation
(FG) college students (Harackiewicz et al., 2014), and
women (Miyake et al., 2010; Walton et al., 2015). For
instance, Brady et al. (2016) explored the long-term effects
of a VA intervention. In the intervention condition, students
picked one option from a preselected list of personal values
and wrote about why that value was important on a personal
level. Two years later, the same students were recruited for a
follow-up. Interestingly, their findings showed that the racial
achievement gap among Latinx students was reduced and
their grades increased. Brady et al. attributed these results to
more self-affirming and less self-threatening thoughts and
feelings in response to adversity in school. The VA interven-
tion has also been shown to be beneficial for women in
STEM. Walton et al. (2015) explored the VA intervention
with women in engineering found similarly positive effects.
Specifically, self-affirmation helped women in engineering
improve their academic attitudes as well as their GPAs.
Interestingly, Walton et al. found that women who self-
affirmed developed stronger gender identification, experi-
enced less threat, and performed on par with their male peers
on mathematics tests.
Dowell et al.
4
Most relevant for the current research is Miyake et al.’s
(2010) study, which tested the effectiveness of a VA inter-
vention with women in an undergraduate physics course.
The study was a randomized, double-blind study in which
students were assigned to write about their most important
(intervention group) or least important (control group) val-
ues two times during the course. Miyake et al. found that
female students in the intervention group improved their
course grade by a full letter grade, on average, and improved
their scores on a standardized physics test.
While encouraging, the positive results of the VA inter-
vention have been largely limited to implementations within
a single classroom or lab experiment. Efforts to move
beyond a boutique remedy and close achievement gaps for
large numbers of students have been inconsistent at best
(Borman et al., 2018; Serra-Garcia et al., 2020), and unsuc-
cessful at worst, highlighting the potential fragility of the VA
in educational settings at scale, and the need for new quanti-
fiable measures and evidence regarding the necessary condi-
tions for effective VA interventions (Hanselman et al., 2017).
Implementation fidelity and intervention processes have
been used as a way to explain the inconsistencies of results
(Bradley et al., 2015; Yeager & Walton, 2011). These inves-
tigations have primarily focused on external and static fea-
tures such as implementation and delivery details (e.g.,
timing of the intervention, manner in which the intervention
is framed) and contextual conditions (e.g., location of the
writing—identity threats “in the air” in a particular setting).
However, there is a comparatively limited body of research
that has explored the actual content of student’s essays (e.g.,
Tibbetts et al., 2016). This is surprising given that the
dynamic, cognitive, and psychological mechanisms are
externalized in the language and discourse features that
characterize students’ VA essays. The current research
addresses this gap by leveraging automated NLP and com-
putational modeling to characterize the linguistic features of
students’ VA essays which are related to more or less benefi-
cial outcomes.
Text as Data: Linguistic Analysis in Psychological
Interventions
Student-generated written responses are a critical compo-
nent of many psychological interventions, including the val-
ues affirmation intervention (Akcaoglu et al., 2018; Riddle
et al., 2015). Students’ essays produced during such inter-
ventions can provide a valuable window into the processes
that may contribute (more or less) to the beneficial effect of
interventions. However, to date, there has been only a hand-
ful of studies that have investigated the language and dis-
course features underlying psychological interventions more
broadly (Harackiewicz et al., 2014; Klebanov et al., 2017),
and the VA intervention in particular (Hanselman et al.,
2017; Shnabel et al., 2013; Tibbetts et al., 2016).
Some researchers have relied on more conventional
approaches that require human examination (i.e., manual
content analysis; Krippendorff, 2003) to characterize the
content of student’s intervention essays (e.g., Borman et al.,
2018; Hanselman et al., 2017; Harackiewicz et al., 2016;
Shnabel et al., 2013). Many of these studies coded content
toward a goal of manipulation checks, wherein essays were
coded to assess the degree to which they showed evidence of
self-affirming reflection (e.g., Hanselman et al., 2017) or the
level of utility value articulated in an essay (e.g.,
Harackiewicz et al., 2016). While useful, this approach is
not focused on characterizing features of the texts, such as
themes, sentiment, or cohesion. In contrast, Shnabel et al.
(2013) used manual content analysis to qualitatively exam-
ine whether student’s VA essays explicitly articulated their
values as connected to some sense of “social belonging”
(e.g., one values an activity because it is done with others).
Qualitative text analysis approaches can provide useful
information, but are also known to carry biases and other
methodological limitations (Krippendorff, 2004). In particu-
lar, the laborious nature of manual coding essays make them
a less viable option with the increasing scale of data (Crossley
et al., 2019; Dowell, Graesser, et al., 2016; Joksimović et al.,
2018; Li et al., 2018; McNamara et al., 2017).
As such, researchers have been incorporating automated
linguistic analysis, including more shallow-level word
counts and deeper level discourse analysis approaches. Both
levels of linguistic analysis are informative. Content analy-
sis using word-counting methods allows getting a fast over-
view of learners’ participation levels, as well as assessing
specific words and word categories. Advances in artificial
intelligence methods, such as NLP (Kao & Poteet, 2007),
have made it possible to automatically (a) harness vast
amounts of educational discourse data being produced in
technology-mediated learning environments, (b) quantify
aspects of human cognition, affective, and social processes
that (c) would otherwise not be possible or extremely time-
consuming for human coders to capture, given the multifac-
eted characteristics of human discourse. Indeed, NLP and
automated text analysis approaches have proven quite useful
in quantifying and characterizing psychological, affective,
cognitive, and social phenomena from a learner-generated
discourse (Bell et al., 2012; Cade et al., 2014; D’Mello et al.,
2009; D’Mello & Graesser, 2012; Dowell et al., 2017, 2019,
2020; Dowell & Graesser, 2015; Eichstaedt et al., 2018;
Kern et al., 2020; Lin et al., 2020; McNamara et al., 2014;
Schwartz et al., 2013; Tausczik & Pennebaker, 2010;
Zedelius et al., 2019).
In the context of wise interventions, there has been grow-
ing efforts devoted toward exploring the content of student’s
psychological intervention essays using automated linguistic
analysis tools, namely, LIWC (Pennebaker et al., 2015).
While the majority of this research has been conducted in the
contexts of the utility-value intervention paradigm (Akcaoglu
Mechanisms Underlying Values Affirmation Interventions
5
et al., 2018; Harackiewicz et al., 2016; Hecht et al., 2019;
Klebanov et al., 2017, 2018; Priniski et al., 2019), there have
been a notable few devoted toward understanding the lin-
guistic mechanisms underlying student’s values affirmation
intervention essays (Riddle et al., 2015; Tibbetts et al.,
2016). For instance, Tibbetts et al. (2016) conducted a fol-
low-up study of the Harackiewicz et al. (2014) sample and
found that the VA intervention was beneficial for FG stu-
dents’ overall postintervention GPAs over the course of a
3-year period. Particularly, relevant to the current research,
they used LIWC to automatically quantify the degree to
which student’s essays exhibited independent and interde-
pendent individual orientations. Words included in the inde-
pendent dictionary included themes of individual interest
and achievement, self-discovery, uniqueness, and leader-
ship. Words in the interdependent dictionary reflected inter-
personal themes of belonging, family, support, and empathy.
They found that the effects of the VA intervention on course
grades, academic belonging, and overall GPA 3 years later
were all mediated by independent themes. In other words,
for FG students, writing about independence in their VA
essays led to higher grades in the biology course, higher lev-
els of academic belonging, and higher GPAs over a 3-year
period (Harackiewicz & Priniski, 2018).
As evident from this research, language can provide a
powerful and measurable behavioral signal that can be used
to capture the semantic processes and psychological con-
structs elicited during psychological interventions, and offer
new insights into how different groups internalize interven-
tion messages, and the linguistic mechanisms that incur the
greatest benefits for students (Harackiewicz & Priniski,
2018; Hecht et al., 2019; Priniski et al., 2019). In the current
research, we explored student’s VA intervention essays using
two well-established and complementary automated text
analysis tools, namely, LIWC (Pennebaker et al., 2015) and
Coh-Metrix (Graesser et al., 2004; McNamara et al., 2014;
see Method section for more details).
This novel combination allowed us to quantify both psy-
chologically meaningful word categories (i.e., LIWC) and
discourse elements (i.e., cohesion; Coh-Metrix). In particu-
lar, LIWC allows us to quantify constructs directly relevant
to the VA intervention, including references to family, inde-
pendent, and interdependent individual orientations (e.g.,
pronouns). Moving past what has been explored previously,
we additionally include constructs that situate these con-
structs within students’ awareness, including temporal orien-
tation, drives, cognition, and sentiment.
Additionally, Coh-Metrix, which employs more sophisti-
cated NLP, allows us to dive deeper and explicitly focus on
the cohesion within students’ essays along two dimensions,
namely, referential and deep cohesion. In line with Kintsch’s
(1998) construction-integration theory, Coh-Metrix distin-
guishes between multiple types of cohesion which fall under
two main forms, namely, textbase (i.e., referential cohesion)
and situation model cohesion (i.e., deep cohesion).
Referential or textbase cohesion is primarily maintained
through the bridging devices, that is, the overlap in words, or
semantic references, whereas deep cohesion related to the
situation model dimension and reflects causation, intention-
ality, space, and time (McNamara et al., 2014). Together this
NLP approach allows us to begin to address the need for
data-driven insights (Paxton & Griffiths, 2017) and research
efforts devoted toward “text analysis of students’ essays may
offer new insights into how different groups internalize
intervention messages and what types of writing interven-
tions have the greatest benefits for students” (Harackiewicz
& Priniski, 2018).
Current Research
To address this need, the current research provides unique
evidence on this issue by reporting the observed differences
between two randomized field implementations of the VA
intervention at scale: (a) a successful traditional in-class
intervention and (b) an unsuccessful online implementation.
The classroom intervention was delivered to 515 students in
an introductory physics course at a Midwestern university
that has experienced considerable GPDs over the years. As
shown in Figure 1, female students in the in-class interven-
tion experienced an increase in both exam grades and their
overall course grade as reported by Koester and McKay
(2021). During the same semester, we implemented an
online VA intervention to 1,936 students across five STEM
courses, which have also experienced considerable GPDs
over the years, using ECoach, a well-established computer-
tailored communication system (Huberth et al., 2015). As
shown in Figure 2, there was no observed improvement in
the GPD for female students in the affirmation condition of
the online intervention.
The first three research questions are aimed at determining
if there is a more successful engagement profile and identify-
ing the language and discourse features that characterize it.
The final research question not only achieves this aim but also
allows us to determine if students’ language and discourse
might be an important factor in the effectiveness of the VA
interventions. Overall, these research questions allowed us to
explore new factors and gain a deeper understanding of the
underlying mechanisms associated with effective VA inter-
ventions for alleviating the GPD in STEM courses.
Research Question 1 (RQ1): Are there unique linguistic
features that differentiate affirmation essays from the
control essays in the classroom intervention (i.e., affir-
mation vs. control)?
Research Question 2 (RQ2): What are the language and
discourse features that differentiate between male and
female VA essays (i.e., not control essays) in the class-
room intervention?
6
Research Question 3 (RQ3): For the classroom interven-
tion, are students’ linguistic profiles associated with
their expected performance?
Research Question 4 (RQ4): What are the language and
discourse features that differentiate between female
students in-class affirmation essays, which was suc-
cessful, and female students’ affirmation essays in the
online intervention, which was not successful?
Method
Participants
In-Class Intervention Participants. A total of 515 stu-
dents enrolled in an electromagnetism-based physics
course participated in the study. Of the 515 students,
two were removed from the analysis due to participant
error. Of the remaining 513 students, 144 were female
and 369 were male. Students were randomly assigned to
receive either the VA intervention (n = 255) or control
(n = 258).
Online Intervention Participants. A total of 1,936 students
across five STEM courses participated in the study. In the
current study, we focus only on the female students’ affirma-
tion essays (n = 538) to address RQ4.
Sample Size. The sample was sufficient to reliably detect
effect sizes (ds) as small as 0.248; 95% confidence interval
(CI) [0.074, 0.421] among VA and control essays (RQ1; n =
513, = .05, 1 − = 0.80), ds as small as 0.352; 95% CI
[0.106, 0.599] among male and female essays (RQ2; n = 255,
= .05, 1 − = 0.80), as small as 0.656; 95% CI = [0.106,
FIGURE 1. Exam grades (A) and final course grade (B) by gender and condition for the in-class VA intervention. Error bars represent
95% confidence intervals. Reprinted with permission from Koester and McKay (2021).
FIGURE 2. Exam grades (A) and final course grade (B) by gender and condition for the online VA intervention. Error bars represent
95% confidence intervals. Reprinted with permission from Koester and McKay (2021).
Mechanisms Underlying Values Affirmation Interventions
7
0.599] among high-performing male and female essays (RQ3;
n = 75, = .05, 1 − = 0.80), and as small as 0.248; 95% CI
[0.074, 0.248] among female students in-class and online affir-
mation essays (RQ4; n = 609, = .05, 1 − = 0.80).
Experimental Procedure: Psychological Interventions
Design and Delivery
Experimental Procedure: In-Class. Students were blocked
by race, gender, cumulative GPA and year in school and
then randomly assigned to either the VA or the control con-
dition. Following randomization, 255 students (71 female)
received the VA exercise while 258 (73 female) students
received the control exercise.
Participants completed the VA writing exercise or control
writing exercise during the lab section of the course. Following
from previously successful VA interventions in comparable
college settings (Harackiewicz et al., 2014; Miyake et al.,
2010) and in accordance with the suggested implementation
standards, the writing exercises took place early in the semes-
ter (i.e., Week 3 of the 15-week semester) and preceded any
course exams. Unlike previous VA research conducted in col-
lege settings (Harackiewicz et al., 2014; Miyake et al., 2010),
this study does not include a second writing exercise midway
through the term. While this is a departure from past imple-
mentations, a single dose of the writing exercise has shown to
be sufficient. In the original test of the VA intervention in
middle school classrooms, Cohen et al. (2006) only used a
single dose of the writing exercise administered at the begin-
ning of the year. Moreover, while Miyake and colleagues pro-
vided students the opportunity to complete the writing exercise
a second time in the middle of the term, this second opportu-
nity was optional and administered online, rather than in the
classroom.
In accordance with the procedures used by Harackiewicz
et al. (2014), the writing exercise was administered by TAs
in the weekly lab section of the course. TAs were naive to the
purpose of the study and were blinded to the students’ condi-
tion. During Week 3 of the semester, a member of the
research team reported to the lab prior to the start of class
and handed the TAs a packet of manila envelopes labeled
with students names. Within each packet was either the VA
writing exercise or the control condition writing exercise
predetermined for each student. TAs also received a stan-
dardized script to introduce the exercise to their students
(see the online Supplemental Material for more details on
the script).
After reading aloud the instructions, TAs proceeded to
pass the labeled envelopes to corresponding students. While
the envelope contained one of two different writing exer-
cises, the exercises closely resembled one another in size
and appearance. In both conditions, the envelope contained
a two-page packet with a list of 14 values on the front of the
first page. The list of values closely resembled those used in
previous college VA interventions (Harackiewicz et al.,
2014; Miyake et al., 2010). After opening the envelope, the
first page of the packet instructed students in the control con-
dition to mark the two to three values that were the least
important to them and write on the next page why they could
be important to someone else. Conversely, students assigned
to the affirmation condition were instructed to mark the two
to three values that were the most important to them and then
on the following page write why these values were impor-
tant to themselves. Students were given 5 minutes to com-
plete the writing exercise. After the exercise was completed,
students placed their packet back into the original manila
envelope with their name label. The envelopes were then
collected by TAs and then returned to the member of the
research team monitoring from the hallway.
Experimental Procedure: Online Intervention. Again, stu-
dents were blocked by race, gender, cumulative GPA, and
year in school and then randomly assigned to either the VA
or the control condition. Following randomization, 538
female students received the VA exercise. To deliver our
intervention, we used ECoach, a well-established computer-
tailored communication system, already delivering person-
alized feedback, encouragement, and advice to thousands of
students per term (Huberth et al., 2015). Students were
invited to complete a writing exercise within the online plat-
form around Week 3 of the semester and preceded any course
exams. Some courses offered extra credit for the exercise,
while others did not. Students who agreed to participate
were randomized to receive either the intervention writing
prompt or a control writing prompt. Following the same pro-
cedure as the classroom intervention, students in the control
condition were asked to mark the two to three values that
were the least important to them and write about why they
could be important to someone else. Conversely, students
assigned to the affirmation condition were instructed to mark
the two to three values that were the most important to them
and then write why these values were important to them. In
both conditions, students were required to write for at least 5
minutes.
Performance Measurement
The performance measure used in the current analyses,
for RQ3, was a relative performance measure that has been
referred to as “better than expected” (BTE; Wright et al.,
2014). Unlike more traditional performance measures (e.g.,
course grade), BTE is a relative estimate of student perfor-
mance—whether a student performed better or worse than
expected (Huberth et al., 2015; Matz et al., 2017). Expected
performance, which has been shown to play a key role in
motivation and achievement, is derived from student charac-
teristics such as prior GPA and standardized test scores. In
this approach, a student receiving a C in physics might be
8
considered BTE if peers with a similar background typically
fail. Likewise, a student with a 4.0 GPA receiving her first
B+ (which others might consider a good grade) would have
a performance that is considered to be worse than expected.
Values Affirmation Intervention Essays
Participant essays were processed to quantify individual
linguistic differences. These analyses yielded individual
summary measures of the engagement and the nature of par-
ticipation. Table 1 reports the descriptive statistics for aver-
age words written and average sentence length between
conditions (intervention and control), genders, and environ-
ments (online and in-class).
A number of conclusions can be drawn from Table 1.
First, the comparisons between student essays constructed
within online versus in-class show that across both condi-
tions, all students (i.e., females and males) wrote substan-
tially more in the online environment, compared with the
in-class. However, the average number of words across both
environments does reflect appropriate task engagement. A
comparison between the intervention and control conditions,
across both environments, shows all students (i.e., females
and males) wrote more in the intervention conditions; how-
ever, the average sentence length was longer in the control
conditions. Finally, a gender comparison shows that females
wrote more than males across both conditions and environ-
ments, but this difference is slightly more pronounced in the
in-class intervention condition.
Computational Evaluation Tools
Prior to computational evaluation, the logs were cleaned
and parsed to facilitate a student-level evaluation. Thus, text
files were created that included each learner’s essay, yield-
ing a total of 513 text files for the in-class intervention and
538 (female essays only) for the online intervention, one for
each student essay. All files were then analyzed using Coh-
Metrix and LIWC.
The linguistic features explored in the current research
were motivated by both related research and because of
potential alignment with the VA intervention.
Coh-Metrix. Coh-Metrix (www.cohmetrix.com) is an auto-
mated linguistics facility that analyzes features of language
and discourse (McNamara et al., 2014). Coh-Metrix incor-
porates automated computational methods of NLP, such as
syntactic parsing and cohesion computation, to capture lan-
guage characteristics at the word-level, sentence-level, and
deeper levels of discourse. Coh-Metrix provides useful
insights into learners’ affective, social, and cognitive pro-
cesses in a variety of digital learning environments (Choi
et al., 2018; D’Mello & Graesser, 2012; Dowell et al., 2014;
Dowell, Graesser, et al., 2016; Graesser et al., 2011; Graesser
et al., 2018; McNamara & Graesser, 2012). Coh-Metrix has
been extensively validated through more than 150 published
studies, which have demonstrated that Coh-Metrix indices
can be used to detect subtle differences in text and discourse
(Graesser, 2011; Graesser et al., 2011; McNamara et al.,
2006; McNamara et al., 2014). In the current research, we
were particularly interested in utilizing Coh-Metrix to quan-
tify properties of cohesion in students’ VA essays. The two
Coh-Metrix cohesion measures used in the current investi-
gation are briefly described:
Deep Cohesion. The extent to which the ideas in the
text are cohesively connected at a deeper conceptual
level that signifies causality or intentionality.
Referential Cohesion. The extent to which explicit
words and ideas in the text are connected with each
other as the text unfolds.
Linguistic Inquiry Word Count. LIWC is an automated text
analysis tool designed for studying the various emotional,
cognitive, structural, and process components present in text
(Pennebaker et al., 2015). As a word-count program, the key
component of LIWC is its embedded dictionary. LIWC pro-
cesses individual or multiple textual files by searching and
TABLE 1
Linguistic Descriptive Statistics for Student Essays Across Intervention Conditions, Environment, and Gender
Affirmation Control
Females Males Females Males
Measure MSEMSEMSEMSE
Online
Mean word count 165.22 2.91 158.75 2.99 139.47 2.69 135.74 2.65
Mean words per sentence 19.66 0.24 19.29 0.26 20.79 0.24 21.23 0.48
In-class
Mean word count 79.68 2.32 65.61 1.58 67.04 2.14 61.61 1.45
Mean words per sentence 18.89 0.60 17.77 0.41 20.63 0.82 19.33 0.44
Mechanisms Underlying Values Affirmation Interventions
9
counting words that are listed in the predesignated diction-
ary. The dictionary itself has been revised and validated over
the course of two decades, and the most recent version con-
sists of 6,400 English words/word stems, covering a range of
social and psychological constructs such as affect, cognition,
and biological processes (see Pennebaker et al., 2015, for
details). Currently, LIWC is one of the most popular and
reliable programs for text analysis available; it has been uti-
lized in hundreds of studies across the social sciences,
including psychology, education, sociology, communica-
tion, political sciences, and economics (Borowiecki, 2017;
Boyd et al., 2020; Cade et al., 2014; Dowell, Windsor, et al.,
2016; Kacewicz et al., 2014; Lin et al., 2020; Newman et al.,
2008; Pennebaker & Chung, 2014; Pennebaker et al., 2014).
A total of 29 linguistic variables from six LIWC categories
were included in the analysis. The LIWC categories used in
the current investigation are briefly described below, and a
full list of the associated 29 LIWC variables can be found in
the online Supplemental Material:
Affective Processes: Words expressing positive and
negative affect, such as love, nice, sweet and hurt,
ugly, nasty, respectively.
Cognitive Processes: Words suggestive of individuals
organizing and intellectually understanding the issues
addressed in their writing (e.g., because, would,
maybe, but).
Pronouns: Words indicating attentional focus such as
I, we, they.
Temporal Focus: Words expressing temporal focus,
including past focus (e.g., ago, did, talked), present
focus (e.g., today, is, now), and future focus (e.g.,
may, will, soon).
Drives: Words expressing an individual’s motiva-
tions, including power, affiliation, and achievement.
Family: Words indicating family relationships (e.g.,
mother, sister, aunt).
Statistical Analyses
Principal Component Analysis
In-class intervention. A PCA approach was adopted
to discover language and discourse patterns associated
with students’ VA and control essays. PCA is a common
data mining technique that involves reducing multidimen-
sional data sets to lower dimensions for analysis (Tabach-
nick & Fidell, 2007). In the current research, it was used
to reduce the 31 linguistic features (two Coh-Metrix indices
and 29 LIWC indices) to create meaningful, broader vari-
ables with which to describe the students’ VA intervention
essays. PCA has been applied in previous studies of psycho-
logical interventions and has proven useful in building an
understanding of language characteristics in student essays
and discourse more broadly (Cade et al., 2014; Dowell &
Graesser, 2015; Pennebaker et al., 2014). Prior to analysis,
the data were normalized, centered, and checked for factor-
ability (for more details, see online Supplemental Material).
The loadings, which quantify the strength of the relationship
between the component and each linguistic variable, were
used to describe and name each component. Table 2 pro-
vides a description of the 10 principal components. Due to
word limit constraints, we are not able to provide illustrative
examples here; however, example student essays from the
current data are provided in the online Supplemental Mate-
rial as an illustrative example of the linguistic features that
comprise a few of the component scores.
Women in-class and online intervention. A separate
PCA was conducted to create meaningful, broader vari-
ables with which to describe the female students’ VA
intervention essays written in the in-class and online inter-
vention. The same procedure was followed as in the pre-
vious analysis, including data normalization, centering,
and factorability evaluation (for more details, see online
Supplemental Material). The loadings, which quantify the
strength of the relationship between the component and
each linguistic variable, were used to describe and name
each component. Table 3 provides a description of the 10
principal components.
Generalized Logistic Mixed-Effects Regressions. A general-
ized logistic mixed-effects modeling approach was adopted
for all analyses due to the structure of the data (e.g., interin-
dividual word count variability; Baayen et al., 2008). Mixed-
effects models include a combination of fixed and random
effects that assess the influence of the fixed effects on depen-
dent variables after accounting for any extraneous random
effects. Mixed-effect modeling provides a robust and flexi-
ble approach that allows for a wide set of correlation patterns
to be modeled.
The analyses for the in-class intervention consisted of
testing for linguistic differences in VA intervention essays
(affirmation vs. control; RQ1) and between males’ and
females’ affirmation essays (RQ2 and RQ3). There were two
sets of dependent measures for the in-class intervention anal-
yses: (a) essay type (affirmation vs. control) and (b) gender.
For RQ2, gender was the dependent variable, and we focused
on the language and discourse features that differentiate
between male and female VA essays only (i.e., not control
essays) in the classroom intervention. This analysis was
motivated to investigate any potential influence of gender on
essay construction, wherein it could be possible that observed
differences on RQ1 were due to males and females construct-
ing essays in a similar fashion. For RQ3, gender was also the
dependent variable; however, learners were grouped into per-
formance bins (i.e., based on a quartile split of their BTE
scores) to investigate any potential influence of performance
on essay construction. Across all models, the independent
10
fixed-effect variables consisted of the 10 linguistic dimen-
sions (i.e., principal components, Table 2).
The analysis for RQ4 consisted of testing for linguistic
differences in VA intervention essays between women who
performed the intervention in-class (i.e., successful), com-
pared with those who performed it online (i.e., unsuccess-
ful). The dependent variable for this analysis was
environment (in-class vs. online), independent fixed-effect
variables consisted of the 10 linguistic dimensions (Table 3).
In addition to constructing the models with the 10 dis-
course features as fixed effects, null models with the random
effects (learner) but no fixed effects were also constructed.
A comparison of the null, random effects only model with
the fixed-effect models allows us to determine whether dis-
course predicts essay type, gender, and environment above
and beyond the random effects (i.e., individual differences in
learners). Akaike information criterion (AIC), log likelihood
(LL), and a likelihood ratio test were used to determine the
best fitting and most parsimonious model. Additionally, the
effect sizes (R2) for each model were estimated according to
Nagelkerke (1991) and Cragg and Uhler’s (1970) pseudo-R2
statistic. The generalized logistic mixed-effects regression
models were conducted using R Version 3.0.1 software for
statistical analysis.
TABLE 2
Description of Principal Components (PCs) for In-Class Intervention
Measure Language Characteristics
Intrapersonal family (PC1) Very intrapersonal focus (I), family references, with less future focused and tentative
language (maybe, perhaps, cognitive processes)
Positive emotion and affiliation (PC2) Positive emotion, affiliation (e.g., social connected references “care,” “help,” “intimate,”
“kind,” “neighbor,” and “volunteer”), and reward drives
Sad anxious (PC3) Very negative, sad, and anxious language
Achievement (PC4) Reward and achievement orientation (take, prize, benefit), work and deep cohesion
Negative past focus (PC5) Less positive emotion, more complex (longer) past-oriented language, and much less present
focused, and certainty
Cohesive future self (PC6) Referential cohesion, more “I” references, and less “we” and more future focused
Tentative future focus (PC7) Very tentative, relative, and future-focused language
Differentiation (PC8) High differentiation (hasn’t, but, else) language
Uncertain perceptions (PC9) Visual perception and uncertain language
Present family cohesion (PC10) Deep and referential cohesion coupled with present focus and family references
TABLE 3
Description of Principal Components (PCs) for Women in the Intervention (In-Class and Online)
Measure Language Characteristics
Positive emotion and affiliation focus (PC1) Positive emotion, affiliation (e.g., social connected references “care,” “help,”
“intimate,” “kind,” “neighbor,” and “volunteer”), and reward drives
Confident intrapersonal family focus (PC2) Very intrapersonal focus (I), family references, with less future focused and less
tentative language (maybe, perhaps, cognitive processes)
Sad and anxious affiliation focus (PC3) Very negative, sad, and anxious language, coupled with affiliation references (e.g.,
social connected references)
Achievement (PC4) Reward and achieve orientation (take, prize, benefit), work and deep cohesion
Interpersonal and future focus (PC5) Interpersonal references (“we”, affiliation), more future focused, and contemplative
orientation (discrepancy—should, would)
Present and future personal achievement (PC6) Achievement and work references, cohesive language, more “I” references, and more
future/present focused
Perceptions (PC7) Perceptual oriented language
Uncertain differentiation (PC8) High differentiation (hasn’t, but, else), and low certainty (always, never) language,
coupled with slightly intrapersonal references (I)
Positive and certain (PC9) Positive emotion, certain, relative (then, finally, after) and a time references (never,
always, end, until)
Deep cohesion (PC10) High levels of deep cohesion, more complex words, adverbs, and a power reward
orientation
11
Results
The likelihood ratio test indicated that the full model was
the most parsimonious and best fit for Intervention Condition
(RQ1), and Affirmation Gender (RQ2) models with 2(1) =
478.03, p < .001, R2 = .81, and 2(1) = 53.28, p < .001, R2
= .27, respectively. A number of conclusions can be drawn
from this initial model fit evaluation and inspection of R2
variance. First, the model comparisons suggest that the dis-
course features were able to add a significant improvement
in differentiating between the learners’ VA and control
essays and between males’ and females’ construction of VA
essays. Second, linguistic characteristics explained about
81% and 27% of the predictable variance in essay type and
gender, respectively. The linguistic characteristics that were
predictive of Intervention Condition type and Gender are
presented in Figures 3 and 4. The reference group was the
Control Essays and Males—meaning that higher odds ratio
indicates higher probability of being a VA Essay or Female
student for each model, respectively.
Table 4 shows the coefficients for the discourse features
that successfully differentiated VA essays from controls,
and female students’ affirmation essays from male stu-
dents’ affirmation essays. As shown in Table 4 and Figure
3, for the Intervention Condition Model, affirmation essays
were characterized by Intrapersonal Family Focus, Positive
Emotion and Affiliation Focus, Achievement, Sad and
Anxious Discourse. However, results also indicate the
opposite association between Uncertain Perceptions,
Tentative Future Focus, Negative Past Focus, and
Differentiation Language with the predicted probability of
being a VA Essay.
The Affirmation Gender Model, explored linguistic dif-
ferences in male and female students’ VA essays for the in-
class intervention (RQ2). The Affirmation Gender Model
(Figure 4) shows that female students’ affirmation essays,
compared with males’, were characterized more by
Intrapersonal Family Focus, and Present Family Cohesion.
Compared with males, female student’s affirmation essays
also used significantly less Sad and Anxious Discourse,
Tentative Future Focus, Negative Past Focus language.
It is possible that the observed linguistic differences in
male and female students VA essays are simply a product of
performing similarly within the class (RQ3). Thus, the third
set of analyses involved a more fine-grained investigation of
how higher and lower performing males and females con-
structed their essays (both VA and control). In order to
explore higher and lower performing students, we created
three bins of learners based on a quartile split of their BTE
scores. This resulted in roughly 75 learners per bin. The
lower and higher bins were used for analysis while the mid-
dle was excluded to reduce noise. Four models were con-
structed, where two models were VA essays for higher and
lower performing students, and two models were control
essays for higher and lower performing students. Particularly,
for both conditions (i.e., VA and control), we constructed a
higher BTE model, and a lower BTE model. For all models,
the linguistic characteristics were the independent vari-
ables, and gender (i.e., male or female) was the dependent
variable.
For the VA analyses, the likelihood ratio test indicated
that the full model was the most parsimonious and best fit
for VA higher BTE model, with 2(10) = 22.76, p < .01, R2
= .46, but not the VA lower BTE model with 2(10) = 15.87,
p = .10. Interestingly, when these relationships were further
explored in the control essay analyses, the likelihood ratio
tests indicated that the full models for control higher BTE
0.40 ***
0.46 ***
0.60 ***
0.72 *
0.95
1.10
1.34 *
1.77 ***
2.21 ***
7.85 ***
Differentiation
Negative Past Focus
Tentative Future Focus
Uncertain Perceptions
Cohesive Future Self
Present Family Cohesion
Sad & Anxious Discourse
Achievement
Positive Emotion &
Affiliation Focus
Intrapersonal Family
Focus
0.20.5 12
51
0
Odds Ratios
Values Affirmation Essay
FIGURE 3. Odds ratios for intervention condition model. Error
bars represent 95% confidence intervals. The reference group
was the control essays, meaning that higher odds ratio indicates
higher probability of being a values affirmation essay.
*p < .05. **p < .01. ***p < .001.
0.60 ***
0.60 **
0.65 **
0.85
0.86
0.96
1.02
1.11
1.47 **
1.56 **
Negative Past Focus
Tentative Future Focus
Sad & Anxious Discourse
Achievement
Positive Emotion &
Affiliation Focus
Cohesive Future Self
Differentiation
Uncertain Perceptions
Present Family Cohesion
Intrapersonal Family
Focus
12
Odds Ratios
Gender
FIGURE 4. Odds ratios for affirmation gender model. Error
bars represent 95% confidence intervals. The reference group
was males, meaning that higher odds ratio indicates higher
probability of being a female student essay.
*p < .05. **p < .01. ***p < .001.
12
and control lower BTE model did not yield a significantly
better fit than the null model with 2(10) = 4.71, p = .90,
and 2(10) = 9.66, p = .47, respectively.
The model comparisons suggest that the discourse fea-
tures were able to add a significant improvement in differen-
tiating between the higher performing male and female
learners’ VA essays, however, no discerning difference was
observed between lower performing male and female learn-
ers’ VA essays, or higher and lower performing learners’
control essays. This suggests that good essays may be an
important mechanism for effective VA interventions. More
specifically, this indicates that perhaps how students con-
struct their essays in terms of linguistic characteristics may
be an important construct in the underlying mechanisms
driving the beneficial effect of the intervention for women.
Second, the linguistic features of high-performing learners’
essays explained about 46% of the predictable variance in
gender differences. The linguistic characteristics that signifi-
cantly discriminate between high-performing male and
female essays are presented in Figure 5 with the odds ratios
and confidence intervals. The reference group was Males,
meaning that higher odds ratio indicates higher probability
of being a female student’s essay. As highlighted in Figure 5,
we observed a significant effect for Intrapersonal Family
Focus ( = 1.19, SE = 0.58, p < .05), and a marginally
significant effect for Sad and Anxious Discourse ( = 1.01,
SE = 0.51, p = .05).
The final analysis focused on investigating the language
and discourse features that characterize female learners’ VA
essays in-class (i.e., successful) and in the online version
(i.e., unsuccessful) of the intervention. Here, we constructed
one model, Women Environment Model (RQ4), with environ-
ment (in-class vs. online) as the dependent variable, the 10
linguistic dimensions as the independent variables, and par-
ticipant as the random effect. The likelihood ratio test indi-
cated that the full model was the best fit for the data with
2(10) = 84.95, p < .001, R2 = .25. The linguistic character-
istics that significantly discriminated between females in the
classroom intervention and online intervention are presented
in Figure 6 with the odds ratio and confidence intervals. The
reference group was online female students’ essays, meaning
that higher odds ratio indicates higher probability of being a
female student’s essay in the classroom intervention. As
highlighted in Figure 6, we observed a significant difference
TABLE 4
Mixed-Effects Model Coefficients for Predicting Intervention Condition Type, and Gender With Language Characteristics
Measure
Intervention Condition Affirmation Gender
SE SE
Intrapersonal family 2.06*** 0.20 0.45* 0.18
Positive emotion and affiliation 0.79*** 0.13 −0.15 0.11
Sad anxious 0.29* 0.13 −0.44** 0.16
Achievement 0.57*** 0.13 −0.16 0.11
Negative past focus −0.78*** 0.15 −0.52** 0.17
Cohesive future self −0.05 0.14 −0.04 0.14
Tentative future focus −0.51*** 0.15 −0.52** 0.19
Differentiation −0.91*** 0.18 0.02 0.16
Uncertain perceptions −0.33*0.14 0.10 0.17
Present family cohesion 0.10 0.15 0.39*0.16
Note. Intervention Condition Model N = 513, Affirmation Gender Model N = 255. The reference group was the Control Essays and Males, meaning that
higher odds ratio indicates higher probability of being a VA Essay or Female student for each model, respectively. = Fixed-effect coefficient; SE = stan-
dard error; VA = values affirmation.
*p < .05. **p < .01. ***p < .001.
3.29 *
1.44
0.36
0.86
0.52
1.37
0.96
1.24
0.49
1.01
Sad&Anxious Discourse
Uncertain Perceptions
Negative Past Focus
Achievement
Tentative Future Focus
Present Family Cohesion
Differentiation
Cohesive Future Self
Positive
Emotion&Affiliation Focus
Intrapersonal Family
Focus
0.05 0.10.2 0.51 2510 20
Odds Ratios
FIGURE 5. Odds ratios and 95% confidence intervals for values
affirmation higher “better than expected” model. The reference
group was males, meaning that higher odds ratio indicates higher
probability of being a female student’s essay.
*p < .05. **p < .01. ***p < .001.
Mechanisms Underlying Values Affirmation Interventions
13
for Positive Emotion and Affiliation Focus ( = 0.76, SE =
0.35, p < .05), and Sad & Anxious Affiliation Focus ( =
−0.76, SE = 0.37, p < .05). Additionally, there were several
linguistic dimensions that were marginally significant,
namely Achievement ( = −0.52, SE = 0.29, p = .07),
Uncertain Differentiation ( = −0.50, SE = 0.30, p = .09)
and Positive and Certain Language ( = 0.58, SE = 0.34, p
= .09). As noted in Koester and McKay (2021), female learn-
ers in the online version of the intervention did not experi-
ence the same performance change as the female students in
the classroom intervention. Had we observed a similar lin-
guistic pattern for both female populations, this might lend
evidence toward the context hypothesis. However, as it
stands, the results suggest that what might be more important
is how individuals construct the VA essays, than where they
construct the essays.
Discussion
There have been attempts to use VA interventions to
address gender-based achievement gaps (e.g., Miyake et al.,
2010), identify the linguistic features associated with its
beneficial effects (Tibbetts et al., 2016), and most recently
scale the intervention (Borman et al., 2018). However, to our
knowledge, this is the first study designed to alleviate gen-
der-based achievement gaps that has also attempted to scale
the intervention and disentangle the beneficial psychological
constructs elicited, from a linguistic point of view, between
classroom and online implementations. Specifically, we
explored the extent to which characteristics of discourse
diagnostically reveal the unique linguistic profile associated
with students’ VA essays, and in particular, more and less
successful VA intervention essays. The findings present
some methodological and theoretical implications for both
intervention scientists and teachers. First, as a methodologi-
cal contribution, we have highlighted the rich contextual
information that can be garnered from using NLP techniques
to reveal more proximal underlying intervention mecha-
nisms. Indeed, NLP has been advocated for in the literature
(Harackiewicz & Priniski, 2018) as a means to gain addi-
tional insights into how different groups internalize inter-
vention messages and what types of writing interventions
have the greatest benefits for students. Particularly, in the
current study, students’ discourse features added significant
improvement in predicting the essential characteristics of
the intervention including the essay type, gender, and inter-
vention context.
We first established that there was a distinct linguistic pro-
file that distinguished VA essays from control essays, and
then explored the linguistic differences in how female and
male students constructed their VA essays (i.e., Affirmation
Gender Model). Our results here were somewhat contradic-
tory to previous research (Tibbetts et al., 2016). Tibbetts et al.
(2016) used LIWC to analyze FG students’ VA essays and
found that students who employed more linguistic features
associated with independence in their VA essays, rather than
interdependence, led to higher grades in their biology course.
Shnabel et al. (2013) identified social belonging (i.e., writing
that reminds students of their interdependence) as the mecha-
nism facilitating the positive effects of VA for Black middle
school students. In contrast, we identified, among others, that
a mix of independent and interdependent language (e.g.,
intrapersonal family and positive affiliation focus), was a
potential mechanism driving the positive effects for female
students. This discrepancy in findings highlights the impor-
tance of giving careful consideration to the target population
before transferring VA strategies, and clearly demonstrates
the need for additional research to understand how these lin-
guistic constructs operate across different groups.
We investigated whether the observed difference between
male and female students’ essay construction was a product
of similar performance (i.e., the high- and low-performing
student models). A noteworthy finding from this concerns
the fact that, when we grouped students by performance, we
only observed a difference between high-performing male
and female students’ VA essay construction, however, no dif-
ference was detected for the other groups (low VA, low and
high control). This provides confidence that the observed
linguistic differences are not simply a product of students
being high performers, but instead offers evidence suggest-
ing that how students construct their essays in terms of lin-
guistic characteristics may be an important construct in the
underlying mechanisms driving the beneficial effect of the
intervention for women.
Our final analysis sought to gain insight into whether lan-
guage and discourse might help explain why female learners
2.13 *
1.28
0.47 *
0.59
1.00
0.95
0.95
0.61
1.79
1.24
Sad&Anxious Affiliation
Discourse
Achievement
Uncertain Differentiation
Present&Future Personal
Achievement
Perceptions
Interpersonal&Future
Focus
Deep Cohesion
Confident Intrapersonal
Family Focus
Positive and Certain
Language
Positive
Emotion&Affiliation Focus
0.10.2 0.51 2510
Odds Ratios
FIGURE 6. Odds ratios and 95% confidence intervals for the
women environment model. The reference group for this analysis
is women in the online intervention, with higher odds indicating
an increased likelihood of it being a female in-class.
*p < .05. **p < .01. ***p < .001.
Dowell et al.
14
who completed the intervention online did not experience
the same performance change as the female students in the
classroom intervention. If female students wrote similarly in
both environments, this would provide evidence in favor of
the social context hypothesis (Steele, 1997), which states
that the effectiveness of self-affirmation approaches depends
on the identity threats “in the air” in a particular setting (i.e.,
the classroom). However, as a theoretical contribution, our
results suggest that what might be more important is how
individuals construct the VA essays, than where they con-
struct the essays. It is important to note that the findings are
not a product of females in class simply being more prolific,
because students actually wrote more in the online interven-
tion. We cannot entirely rule out the social context hypothe-
sis, however, but our results do suggest that there are at least
other important elements at play. In our future research, we
will be designing a scaled online intervention specifically
geared toward eliciting the identified linguistic features from
students writing. Additionally, we plan to use a causal mod-
eling approach to identify actual causal relationships
between specific linguistic features in VA essays and benefi-
cial educational outcomes.
While the automated text analysis approaches utilized in
the current research provide several opportunities, applying
these methodological approaches to real-world data brings
new risks (Iliev et al., 2015; Schwartz & Ungar, 2015). For
instance, word-count-based methods, such as LIWC, lack the
contextual information that is available with human judg-
ment. An interesting illustrative example of this was high-
lighted in the Back et al. (2011) work that explored the
emotional content of text messages sent in the aftermath of
September 11, 2001. A notable finding of this work was that
the timeline of anger-related words showed an intense trend
that kept constantly increasing for several hours after the
attack. However, revisiting this research showed that many of
the text messages were automatically generated by phone
servers (“critical” server problem), and although unrelated to
the theoretical question, they were identified as anger-related
words by the system (Pury, 2011). This cautionary tale high-
lights the need for careful consideration when utilizing auto-
mated text analytic approaches with real-world data.
Despite these limitations, the present research does help
advance our understanding of the VA intervention by high-
lighting critical language and discourse features that qualify
VA effectiveness in buffering against identity threat with the
potential to alleviate GPDs in STEM courses. In doing so, it
furthers research beyond knowing that it can work in one
context, to understand how to potentially make it work in
educational settings at scale and close achievement gaps for
larger numbers of students. Overall, this work helps inform
affirmation theory by suggesting that the processes set in
motion through self-affirmation interventions, for women in
STEM, may be facilitated when these interventions involve
specific language and discourse features.
Acknowledgments
This work was partially supported by the National Science
Foundation 15-585, National Science Foundation No. 1535300, and
by the office of Academic Innovation, and the Holistic Modeling of
Education project funded by the Michigan Institute of Data Science.
ORCID iDs
Nia M. M. Dowell https://orcid.org/0000-0002-9839-8947
Timothy A. McKay https://orcid.org/0000-0001-9036-6150
References
Akcaoglu, M., Rosenberg, J. M., Ranellucci, J., & Schwarz, C.
V. (2018). Outcomes from a self-generated utility value inter-
vention on fifth and sixth-grade students’ value and interest in
science. International Journal of Educational Research, 87,
67–77. https://doi.org/10.1016/j.ijer.2017.12.001
Aronson, J., Cohen, G., & Nail, P. R. (1999). Self-affirmation theory:
An update and appraisal. In E. Harmon-Jones, & J. Mills (Eds.),
Cognitive dissonance: Progress on a pivotal theory in social
psychology (Vol. 2, pp. 159–174). American Psychological
Association. https://doi.org/10.1037/0000135-008
Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-
effects modeling with crossed random effects for subjects and
items. Journal of Memory and Language, 59(4), 390–412.
https://doi.org/10.1016/j.jml.2007.12.005
Back, M. D., Küfner, A. C., & Egloff, B. (2011). “Automatic or
the people?” Anger on September 11, 2001, and lessons learned
for the analysis of large digital data sets. Psychological Science,
22(6), 837−838. https://doi.org/10.1177/0956797611409592
Beilock, S. L., Rydell, R. J., & McConnell, A. R. (2007). Stereotype
threat and working memory: Mechanisms, alleviation, and spill-
over. Journal of Experimental Psychology. General, 136(2),
256–276. https://doi.org/10.1037/0096-3445.136.2.256
Bell, C. M., McCarthy, P. M., & McNamara, D. S. (2012). Using
LIWC and Coh-Metrix to investigate gender differences in lin-
guistic styles. In P. M. McCarthy, & C. Boonthum-Denecke
(Eds.), Applied natural language processing: Identification,
investigation and resolution (pp. 545–556). IGI Global. https://
doi.org/10.4018/978-1-60960-741-8.ch032
Borman, G. D., Choi, Y., & Hall, G. J. (2020). The impacts of a
brief middle-school self-affirmation intervention help propel
African American and Latino students through high school.
Journal of Educational Psychology. Advance online publica-
tion. https://doi.org/10.1037/edu0000570
Borman, G. D., Grigg, J., Rozek, C. S., Hanselman, P., & Dewey, N.
A. (2018). Self-affirmation effects are produced by school con-
text, student engagement with the intervention, and time: Lessons
from a district-wide implementation. Psychological Science,
29(11), 1773-1784. https://doi.org/10.1177/0956797618784016
Borowiecki, K. J. (2017). How are you, my dearest Mozart? Well-
being and creativity of three famous composers based on their
letters. Review of Economics and Statistics, 99(4), 591–605.
https://doi.org/10.1162/REST_a_00616
Boyd, R. L., Blackburn, K. G., & Pennebaker, J. W. (2020). The
narrative arc: Revealing core narrative structures through text
analysis. Science Advances, 6(32), Article eaba2196. https://
doi.org/10.1126/sciadv.aba2196
Mechanisms Underlying Values Affirmation Interventions
15
Bradley, D., Crawford, E., & Dahill-Brown, S. E. (2015).
Fidelity of implementation in a large-scale, randomized, field
trial: Identifying the critical components of values affirma-
tion. Proceedings of the Society for Research on Educational
Effectiveness (ED562183). Society for Research on Educational
Effectiveness. https://eric.ed.gov/?id=ED562183
Brady, S. T., Reeves, S. L., Garcia, J., Purdie-Vaughns, V., Cook, J.
E., Taborsky-Barba, S., Tomasetti, S., Davis, E. M., & Cohen, G.
L. (2016). The psychology of the affirmed learner: Spontaneous
self-affirmation in the face of stress. Journal of Educational
Psychology, 108(3), 353–373. https://doi.org/10.1037/edu0000091
Brewe, E., Sawtelle, V., Kramer, L. H., O’Brien, G. E., Rodriguez,
I., & Pamelá, P. (2010). Toward equity through participation in
Modeling Instruction in introductory university physics. Physical
Review Special Topics—Physics Education Research, 6(1), Article
010106. https://doi.org/10.1103/PhysRevSTPER.6.010106
Cade, W. L., Dowell, N. M., Graesser, A. C., Tausczik, Y. R., &
Pennebaker, J. W. (2014). Modeling student socioaffective
responses to group interactions in a collaborative online chat envi-
ronment. In J. Stamper, Z. Pardos, M. Mavrikis, & B. M. McLaren
(Eds.), Proceedings of the Seventh International Conference on
Educational Data Mining (pp. 399–400). Springer.
Casad, B. J., Oyler, D. L., Sullivan, E. T., McClellan, E. M.,
Tierney, D. N., Anderson, D. A., Greeley, P. A., Fague, M.
A., & Flammang, B. J. (2018). Wise psychological interven-
tions to improve gender and racial equality in STEM. Group
Processes & Intergroup Relations, 21(5), 767–787. https://doi.
org/10.1177/1368430218767034
Cheryan, S., Plaut, V. C., Davies, P. G., & Steele, C. M. (2009).
Ambient belonging: How stereotypical cues impact gender par-
ticipation in computer science. Journal of Personality and Social
Psychology, 97(6), 1045–1060. https://doi.org/10.1037/a0016239
Choi, H., Dowell, N. M., & Brooks, C. (2018). Social compari-
son theory as applied to MOOC student writing: Constructs for
opinion and ability. In J. Kay, & R. Luckin (Eds.), Proceedings
of the 13th International Conference for the Learning Sciences
(pp. 1421–1422). International Society of the Learning Sciences.
Cohen, G.L., Garcia, J., Apfel, N., & Master, A. 2006. Reducing
the racial achievement gap: A social-psychological interven-
tion. Science, 313, 1307–1310.
Cohen, G. L., Garcia, J., Purdie-Vaughns, V., Apfel, N., &
Brzustoski, P. (2009). Recursive processes in self-affirmation:
Intervening to close the minority achievement gap. Science,
324(5925), 400–403. https://doi.org/10.1126/science.1170769
Cohen, G. L., & Sherman, D. K. (2014). The psychology of change:
Self-affirmation and social psychological intervention. Annual
Review of Psychology, 65, 333–371. https://doi.org/10.1146/
annurev-psych-010213-115137
Conger, D., & Long, M. C. (2010). Why are men falling behind?
Gender gaps in college performance and persistence. Annals of
the American Academy of Political and Social Science, 627(1),
184–214. https://doi.org/10.1177/0002716209348751
Cragg, S. G., & Uhler, R. (1970). The demand for automobiles.
Canadian Journal of Economics, 3, 386–406. https://doi.
org/10.2307/133656
Creech, L. R., & Sweeder, R. D. (2012). Analysis of student per-
formance in large-enrollment life science courses. CBE Life
Sciences Education, 11(4), 386–391. https://doi.org/10.1187/
cbe.12-02-0019
Crossley, S. A., Kim, M., Allen, L., & McNamara, D. (2019).
Automated summarization evaluation (ASE) using natural lan-
guage processing tools. In S. Isotani, E. Millán, A. Ogan, P.
Hastings, B. McLaren., & R. Luckin (Eds.), Artificial intelli-
gence in education: AIED 2019 (Lecture Notes in Computer
Science, Vol. 11625, pp. 84–95). Springer. https://doi.
org/10.1007/978-3-030-23204-7_8
Dasgupta, N. (2011). Ingroup experts and peers as social vaccines
who inoculate the self-concept: The stereotype inoculation
model. Psychological Inquiry, 22(4), 231–246. https://doi.org/1
0.1080/1047840X.2011.607313
Dasgupta, N., Scircle, M. M., & Hunsinger, M. (2015). Female
peers in small work groups enhance women’s motivation,
verbal participation, and career aspirations in engineering.
Proceedings of the National Academy of Sciences of the United
States of America, 112(16), 4988–4993. https://doi.org/10.1073/
pnas.1422822112
Dasgupta, N., & Stout, J. G. (2014). Girls and women in science,
technology, engineering, and mathematics: STEMing the tide
and broadening participation in STEM careers. Policy Insights
from the Behavioral and Brain Sciences, 1(1), 21–29. https://
doi.org/10.1177/2372732214549471
D’Mello, S., Dowell, N. M., & Graesser, A. C. (2009). Cohesion
relationships in tutorial dialogue as predictors of affective
states. In V. Dimitrova, R. Mizoguchi, B. DuBoulay, & A.
Graesser (Eds.), Artificial intelligence in education (Vol. 200,
pp. 9–16). IOS Press.
D’Mello, S., & Graesser, A. C. (2012). Language and discourse
are powerful signals of student emotions during tutoring. IEEE
Transactions on Learning Technologies, 5(4), 304–317. https://
doi.org/10.1109/TLT.2012.10
Dowell, N. M., Brooks, C., Kovanović, V., Joksimović, S., &
Gašević, D. (2017). The changing patterns of MOOC discourse.
In C. Urrea, J. Reich, & C. Thille (Eds.), Proceedings of the
Fourth (2017) ACM Conference on Learning @ Scale (pp.
283–286). Association for Computing Machinery. https://doi.
org/10.1145/3051457.3054005
Dowell, N. M., Cade, W. L., Tausczik, Y. R., Pennebaker, J. W.,
& Graesser, A. C. (2014). What works: Creating adaptive and
intelligent systems for collaborative learning support. In S.
Trausan-Matu, K. E. Boyer, M. Crosby, & K. Panourgia (Eds.),
Twelfth International Conference on Intelligent Tutoring
Systems (pp. 124–133). Springer.
Dowell, N. M., & Graesser, A. C. (2015). Modeling learners’ cog-
nitive, affective, and social processes through language and dis-
course. Journal of Learning Analytics, 1(3), 183–186. https://
doi.org/10.18608/jla.2014.13.18
Dowell, N. M., Graesser, A. C., & Cai, Z. (2016). Language and
discourse analysis with Coh-Metrix: Applications from edu-
cational material to learning environments at scale. Journal
of Learning Analytics, 3(3), 72–95. https://doi.org/10.18608/
jla.2016.33.5
Dowell, N. M., Lin, Y., Godfrey, A., & Brooks, C. (2020).
Exploring the relationship between emergent sociocognitive
roles, collaborative problem-solving skills and outcomes: A
group communication analysis. Journal of Learning Analytics,
7(1), 38–57. https://doi.org/10.18608/jla.2020.71.4
Dowell, N. M., Nixon, T., & Graesser, A. C. (2019). Group com-
munication analysis: A computational linguistics approach
Dowell et al.
16
for detecting sociocognitive roles in multi-party interactions.
Behavior Research Methods, 51(3), 1007–1041. https://doi.
org/10.3758/s13428-018-1102-z
Dowell, N. M., Windsor, L. C., & Graesser, A. C. (2016).
Computational linguistics analysis of leaders during crises in
authoritarian regimes. Dynamics of Asymmetric Conflict, 9(1-
3), 1–12. https://doi.org/10.1080/17467586.2015.1038286
Eddy, S. L., & Brownell, S. E. (2016). Beneath the numbers: A
review of gender disparities in undergraduate education across
science, technology, engineering, and math disciplines. Physical
Review Physics Education Research, 12(2), Article 020106.
https://doi.org/10.1103/PhysRevPhysEducRes.12.020106
Eddy, S. L., Brownell, S. E., & Wenderoth, M. P. (2014). Gender
gaps in achievement and participation in multiple introductory
biology classrooms. CBE Life Sciences Education, 13(3), 478–
492. https://doi.org/10.1187/cbe.13-10-0204
Eichstaedt, J. C., Smith, R. J., Merchant, R. M., Ungar, L. H.,
Crutchley, P., Preoţiuc-Pietro, D., Asch, D. A., & Schwartz, H.
A. (2018). Facebook language predicts depression in medical
records. Proceedings of the National Academy of Sciences of
the United States of America, 115(44), 11203–11208. https://
doi.org/10.1073/pnas.1802331115
Fogliati, V. J., & Bussey, K. (2013). Stereotype threat reduces
motivation to improve: Effects of stereotype threat and feed-
back on women’s intentions to improve mathematical ability.
Psychology of Women Quarterly, 37(3), 310–324. https://doi.
org/10.1177/0361684313480045
Forbes, C. E., Schmader, T., & Allen, J. J. B. (2008). The role
of devaluing and discounting in performance monitoring: A
neurophysiological study of minorities under threat. Social
Cognitive and Affective Neuroscience, 3(3), 253–261. https://
doi.org/10.1093/scan/nsn012
Graesser, A. C. (2011). Learning, thinking, and emoting with dis-
course technologies. American Psychologist, 66(8), 746–757.
https://psycnet.apa.org/journals/amp/66/8/746/
Graesser, A. C., Dowell, N., Hampton, A. J., Lippert, A. M., Li,
H., & Williamson, S. D. (2018). Building intelligent conver-
sational tutors and mentors for team collaborative problem
solving: Guidance from the 2015 Program for International
Student Assessment. In Building intelligent tutoring systems for
teams (Vol. 19, pp. 173–211). Emerald. https://doi.org/10.1108/
S1534-085620180000019012
Graesser, A. C., McNamara, D. S., & Kulikowich, J. M. (2011).
Coh-Metrix: Providing multilevel analyses of text character-
istics. Educational Researcher, 40(5), 223–234. https://doi.
org/10.3102/0013189X11413260
Graesser, A. C., McNamara, D. S., Louwerse, M. M., & Cai, Z.
(2004). Coh-metrix: Analysis of text on cohesion and language.
Behavior Research Methods, Instruments, & Computers, 36(2),
193–202. https://www.ncbi.nlm.nih.gov/pubmed/15354684
Hanselman, P., Rozek, C. S., Grigg, J., & Borman, G. D. (2017). New
evidence on self-affirmation effects and theorized sources of het-
erogeneity from large-scale replications. Journal of Educational
Psychology, 109(3), 405–424. https://doi.org/10.1037/edu0000141
Harackiewicz, J. M., Canning, E. A., Tibbetts, Y., Giffen, C. J.,
Blair, S. S., Rouse, D. I., & Hyde, J. S. (2014). Closing the
social class achievement gap for first-generation students in
undergraduate biology. Journal of Educational Psychology,
106(2), 375–389. https://doi.org/10.1037/a0034679
Harackiewicz, J. M., Canning, E. A., Tibbetts, Y., Priniski, S. J.,
& Hyde, J. S. (2016). Closing achievement gaps with a utility-
value intervention: Disentangling race and social class. Journal
of Personality and Social Psychology, 111(5), 745–765. https://
doi.org/10.1037/pspp0000075
Harackiewicz, J. M., & Priniski, S. J. (2018). Improving student
outcomes in higher education: The science of targeted interven-
tion. Annual Review of Psychology, 69(1), 409–435. https://doi.
org/10.1146/annurev-psych-122216-011725
Hecht, C. A., Harackiewicz, J. M., Priniski, S. J., Canning, E. A.,
Tibbetts, Y., & Hyde, J. S. (2019). Promoting persistence in the
biological and medical sciences: An expectancy-value approach
to intervention. Journal of Educational Psychology, 11(8),
1462–1477. https://doi.org/10.1037/edu0000356
Huberth, M., Chen, P., Tritz, J., & McKay, T. A. (2015).
Computer-tailored student support in introductory physics.
PLOS ONE, 10(9), Article e0137001. https://doi.org/10.1371/
journal.pone.0137001
Iliev, R., Dehghani, M., & Sagi, E. (2015). Automated text analysis
in psychology: Methods, applications, and future developments.
Language and Cognition, 7(2), 265–290. https://psycnet.apa.
org/doi/10.1017/langcog.2014.30
Joksimović, S., Dowell, N., Gašević, D., Mirriahi, N., Dawson, S.,
& Graesser, A. C. (2018). Linguistic characteristics of reflective
states in video annotations under different instructional condi-
tions. Computers in Human Behavior, 96, 211–222. https://doi.
org/10.1016/j.chb.2018.03.003
Jordt, H., Eddy, S. L., Brazil, R., Lau, I., Mann, C., Brownell,
S. E., King, K., & Freeman, S. (2017). Values Affirmation
Intervention reduces achievement gap between underrep-
resented minority and White students in introductory biol-
ogy classes. CBE Life Sciences Education, 16(3). https://doi.
org/10.1187/cbe.16-12-0351
Kacewicz, E., Pennebaker, J. W., Davis, M., Jeon, M., & Graesser,
A. C. (2014). Pronoun use reflects standings in social hierar-
chies. Journal of Language and Social Psychology, 33(2), 125–
143. https://doi.org/10.1177/0261927X13502654
Kao, A., & Poteet, S. R. (2007). Natural language processing
and text mining. Springer. https://www.amazon.com/Natural-
Language-Processing-Text-Mining/dp/184628175X
Kern, M. L., Ungar, L. H., & Eichstaedt, J. C. (2020). Estimating
geographic subjective well-being from Twitter: A comparison
of dictionary and data-driven language methods. Proceedings
of the National Academy of Sciences of the United States of
America, 117(19), 10165–10171. https://www.pnas.org/con-
tent/117/19/10165.short
Kintsch, W. (1998). Comprehension: A paradigm for cognition.
Cambridge University Press.
Klebanov, B. B., Burstein, J., Harackiewicz, J. M., Priniski, S. J.,
& Mulholland, M. (2017). Reflective writing about the util-
ity value of science as a tool for increasing STEM motivation
and retention–Can AI help scale up? International Journal of
Artificial Intelligence in Education, 27(4), 791–818. https://doi.
org/10.1007/s40593-017-0141-4
Klebanov, B. B., Priniski, S., Burstein, J., Gyawali, B.,
Harackiewicz, J., & Thoman, D. (2018). Utility-value score:
A case study in system generalization for writing analytics.
Journal of Writing Analytics, 2, 314–328. https://www.ncbi.
nlm.nih.gov/pubmed/31565684
Mechanisms Underlying Values Affirmation Interventions
17
Koester, B. P., Grom, G., & McKay, T. A. (2016). Patterns of gen-
dered performance difference in introductory STEM courses.
arXiv. https://arxiv.org/abs/1608.07565
Koester, B. P., & McKay, T. A. (2021). Gendered performance in
introductory STEM courses [Manuscript submitted for publica-
tion]. Department of Physics, University of Michigan.
Krippendorff, K. (2003). Content analysis: An introduction to its
methodology. Sage.
Krippendorff, K. (2004). Reliability in content analysis. Human
Communication Research, 30(3), 411–433. https://doi.
org/10.1111/j.1468-2958.2004.tb00738.x
Li, H., Cai, Z., & Graesser, A. C. (2018). Computerized sum-
mary scoring: Crowdsourcing-based latent semantic analysis.
Behavior Research Methods, 50(5), 2144–2161. https://doi.
org/10.3758/s13428-017-0982-7
Lin, Y., Yu, R., & Dowell, N. (2020). LIWCs the same, not the
same: Gendered linguistic signals of performance and experience
in online STEM courses. In I. I. Bittencourt, M. Cukurova, K.
Muldner, R. Luckin, & E. Millán (Eds.), Proceedings of the 21st
International Conference: AIED 2020 (Artificial Intelligence
in Education: Part I; Vol. 12163, pp. 333–345). Springer
International. https://doi.org/10.1007/978-3-030-52237-7_27
London, B., Rosenthal, L., & Gonzalez, A. (2011). Assessing the
role of gender rejection sensitivity, identity, and support on the
academic engagement of women in nontraditional fields using
experience sampling methods. Journal of Social Issues, 67(3),
510–530. https://doi.org/10.1111/j.1540-4560.2011.01712.x
Matz, R. L., Koester, B. P., Fiorini, S., Grom, G., Shepard, L.,
Stangor, C. G., Weiner, B., & McKay, T. A. (2017). Patterns of
gendered performance differences in large introductory courses
at five research universities. AERA Open, 3(4). https://doi.
org/10.1177/2332858417743754
McNamara, D. S., Allen, L. K., Crossley, S. A., Dascalu, M., &
Perret, C. A. (2017). Natural language processing and learning
analytics. In C. Lang, G. Siemens, A. F. Wise, & D. Gaevic
(Eds.), Handbook of learning analytics (1st ed., pp. 93–104).
Society for Learning Analytics Research. https://www.solare-
search.org/hla-17/hla17-chapter8/
McNamara, D. S., & Graesser, A. C. (2012). Coh-Metrix: An auto-
mated tool for theoretical and applied natural language process-
ing. In Applied natural language processing: Identification,
investigation and resolution (pp. 188–205). IGI Global. https://
doi.org/10.4018/978-1-60960-741-8.ch011
McNamara, D. S., Graesser, A. C., McCarthy, P. M., & Cai, Z.
(2014). Automated evaluation of text and discourse with Coh-
Metrix. Cambridge University Press.
McNamara, D. S., Ozuru, Y., Graesser, A. C., & Louwerse, M.
(2006). Validating Coh-Metrix. In Proceedings of the 28th
Annual Conference of the Cognitive Science Society (pp. 573–
578). https://www.academia.edu/download/30813803/fpo444-
mcnamara.pdf
McQueen, A., & Klein, W. M. P. (2006). Experimental manipula-
tions of self-affirmation: A systematic review. Self and Identity:
The Journal of the International Society for Self and Identity,
5(4), 289–354. https://doi.org/10.1080/15298860600805325
Miyake, A., Kost-Smith, L. E., Finkelstein, N. D., Pollock, S.
J., Cohen, G. L., & Ito, T. A. (2010). Reducing the gender
achievement gap in college science: A classroom study of val-
ues affirmation. Science, 330(6008), 1234–1237. https://doi.
org/10.1126/science.1195996
Nagelkerke, N. J. D. (1991). A note on a general definition of
the coefficient of determination. Biometrika, 78(3), 691–692.
https://doi.org/10.1093/biomet/78.3.691
Napper, L., Harris, P. R., & Epton, T. (2009). Developing and
testing a self-affirmation manipulation. Self and Identity: The
Journal of the International Society for Self and Identity, 8(1),
45–62. https://doi.org/10.1080/15298860802079786
National Research Council of the National Academies. (2011).
A review of gender differences at critical transitions in the
careers of science, engineering, and mathematics faculty [S.
Bell, Reviewer]. International Journal of Gender, Science and
Technology, 3(1). https://genderandset.open.ac.uk/index.php/
genderandset/article/download/147/249
National Science Board. (2015, February 4). Revisiting the STEM
Workforce, A comparison to science and engineering indicators
2014 (NSB-2015-10). National Science Foundation. https://
www.nsf.gov/pubs/2015/nsb201510/nsb201510
National Science Board. (2016). Developing a National STEM
Workforce strategy: A workshop summary. National Academies
Press. https://www.nap.edu. https://doi.org/10.17226/21900
National Science Foundation. (2019). Women, minorities, and per-
sons with disabilities in science and engineering. https://ncses.
nsf.gov/pubs/nsf19304/
Newman, M. L., Groom, C. J., Handelman, L. D., & Pennebaker,
J. W. (2008). Gender differences in language use: An analysis
of 14,000 text samples. Discourse Processes, 45(3), 211–236.
https://doi.org/10.1080/01638530802073712
Nguyen, H.-H. D., & Ryan, A. M. (2008). Does stereotype threat
affect test performance of minorities and women? A meta-anal-
ysis of experimental evidence. Journal of Applied Psychology,
93(6), 1314–1334. https://doi.org/10.1037/a0012702
Paxton, A., & Griffiths, T. L. (2017). Finding the traces of behav-
ioral and cognitive processes in big data and naturally occur-
ring datasets. Behavior Research Methods, 49(5), 1630–1638.
https://doi.org/10.3758/s13428-017-0874-x
Pennebaker, J. W., Boyd, R. L., Jordan, K., & Blackburn, K. (2015).
The development and psychometric properties of LIWC2015.
University of Texas at Austin. https://repositories.lib.utexas.
edu/handle/2152/31333
Pennebaker, J. W., & Chung, C. K. (2014). Counting little words
in big data: The psychology of individuals, communities, cul-
ture, and history. In J. P. Forgas, O. Vincze, & J. László (Eds.),
Sydney symposium of social psychology. Social cognition and
communication (pp. 25–42). Psychology Press. https://psycnet.
apa.org/fulltext/2013-28261-002.pdf
Pennebaker, J. W., Chung, C. K., Frazee, J., Lavergne, G. M.,
& Beaver, D. I. (2014). When small words foretell academic
success: The case of college admissions essays. PLOS ONE,
9(12), Article e115844. https://doi.org/10.1371/journal.pone.
0115844
Pollock, S. J., Finkelstein, N. D., & Kost, L. E. (2007). Reducing
the gender gap in the physics classroom: How sufficient is
interactive engagement? Physical Review Special Topics—
Physics Education Research, 3(1), Article 010107. https://doi.
org/10.1103/PhysRevSTPER.3.010107
Priniski, S. J., Rosenzweig, E. Q., Canning, E. A., Hecht, C. A.,
Tibbetts, Y., Hyde, J. S., & Harackiewicz, J. M. (2019). The
benefits of combining value for the self and others in utility-
value interventions. Journal of Educational Psychology, 111(8),
1478–1497. https://doi.org/10.1037/edu0000343
Dowell et al.
18
Pury, C. L. (2011). Automation can lead to confounds in text
analysis: Back, Küfner, and Egloff (2010) and the not-so-angry
Americans. Psychological Science, 22(6), 835−836. https://doi.
org/10.1177/0956797611408735
Riddle, T., Bhagavatula, S. S., Guo, W., Muresan, S., Cohen, G.,
Cook, J. E., & Purdie-Vaughns, V. (2015, June 26–29). Mining
a written values affirmation intervention to identify the unique
linguistic features of stigmatized groups. Proceedings of the
Eighth International Conference on Educational Data Mining
(pp. 274–281). International Educational Data Mining Society.
https://eric.ed.gov/?id=ED560575
Schmader, T., & Johns, M. (2003). Converging evidence that ste-
reotype threat reduces working memory capacity. Journal of
Personality and Social Psychology, 85(3), 440–452. https://doi.
org/10.1037/0022-3514.85.3.440
Schmeichel, B. J., & Vohs, K. (2009). Self-affirmation and self-
control: Affirming core values counteracts ego depletion.
Journal of Personality and Social Psychology, 96(4), 770–782.
https://doi.org/10.1037/a0014635
Schwartz, H. A., Eichstaedt, J. C., Kern, M. L., Dziurzynski, L.,
Ramones, S. M., Agrawal, M., Shah, A., Kosinski, M., Stillwell,
D., Seligman, M. E. P., & Ungar, L. H. (2013). Personality, gen-
der, and age in the language of social media: The open-vocab-
ulary approach. PLOS ONE, 8(9), Article e73791. https://doi.
org/10.1371/journal.pone.0073791
Schwartz, H. A., & Ungar, L. H. (2015). Data-driven content
analysis of social media: A systematic overview of automated
methods. Annals of the American Academy of Political and
Social Science, 659(1), 78–94. https://doi.org/10.1177/000271
6215569197
Serra-Garcia, M., Hansen, K. T., & Gneezy, U. (2020). Can
short psychological interventions affect educational perfor-
mance? Revisiting the effect of self-affirmation interven-
tions. Psychological Science, 31(7), 865–872. https://doi.
org/10.1177/0956797620923587
Sherman, D. K. (2013). Self-affirmation: Understanding the
effects. Social and Personality Psychology Compass, 7(11),
834–845. https://doi.org/10.1111/spc3.12072
Sherman, D. K., & Cohen, G. L. (2006). The psychology of
self-defense: Self-affirmation theory. In M. P. Zanna (Ed.),
Advances in experimental social psychology (pp. 183–242).
Elsevier. https://doi.org/10.1016/s0065-2601(06)38004-5
Sherman, D. K., & Hartson, K. A. (2011). Reconciling self-protec-
tion with self-improvement: Self-affirmation theory. In M. D.
Alicke (Ed.), Handbook of self-enhancement and self-protec-
tion (Vol. 524, pp. 128–151). Guilford Press. https://psycnet.
apa.org/fulltext/2011-04015-006.pdf
Sherman, D. K., Hartson, K. A., Binning, K. R., Purdie-Vaughns,
V., Garcia, J., Taborsky-Barba, S., Tomassetti, S., Nussbaum,
A. D., & Cohen, G. L. (2013). Deflecting the trajectory and
changing the narrative: How self-affirmation affects academic
performance and motivation under identity threat. Journal of
Personality and Social Psychology, 104(4), 591–618. https://
doi.org/10.1037/a0031495
Shnabel, N., Purdie-Vaughns, V., Cook, J. E., Garcia, J., & Cohen,
G. L. (2013). Demystifying values-affirmation interventions:
Writing about social belonging is a key to buffering against
identity threat. Personality & Social Psychology Bulletin, 39(5),
663–676. https://doi.org/10.1177/0146167213480816
Skrentny, J., & Lewis, K. (2013). Building the innovation econ-
omy? The challenges of defining, building and maintaining
the STEM Workforce (No. 1). Center for Comparative and
Immigration Studies.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining
the integrity of the self. In L. Berkowitz (Ed.), Advances in experi-
mental social psychology (Vol. 21, pp. 261–302). Academic
Press. https://doi.org/10.1016/S0065-2601(08)60229-4
Steele, C. M. (1997). A threat in the air: How stereotypes shape
intellectual identity and performance. American Psychologist,
52(6), 613–629. https://doi.org/10.1037/0003-066X.52.6.613
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the
intellectual test performance of African Americans. Journal of
Personality and Social Psychology, 69(5), 797–811. https://doi.
org/10.1037/0022-3514.69.5.797
Steele, C. M., Spencer, S. J., & Aronson, J. (2002). Contending
with group image: The psychology of stereotype and social
identity threat. Advances in Experimental Social Psychology,
34, 379–440. https://doi.org/10.1016/S0065-2601(02)80009-0
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate sta-
tistics (5th ed.). Allyn & Bacon/Pearson Education.
Tai, R. H., & Sadler, P. M. (2001). Gender differences in intro-
ductory undergraduate physics performance: University
physics versus college physics in the USA. International
Journal of Science Education, 23(10), 1017–1037. https://doi.
org/10.1080/09500690010025067
Tausczik, Y. R., & Pennebaker, J. W. (2010). The psychological
meaning of words: LIWC and computerized text analysis meth-
ods. Journal of Language and Social Psychology, 29(1), 24–54.
https://doi.org/10.1177/0261927X09351676
Thoman, D. B., & Sansone, C. (2016). Gender bias triggers diverg-
ing science interests between women and men: The role of
activity interest appraisals. Motivation and Emotion, 40(3),
464–477. https://doi.org/10.1007/s11031-016-9550-1
Thoman, D. B., Smith, J. L., Brown, E. R., Chase, J., & Lee, J.
Y. K. (2013). Beyond performance: A motivational experiences
model of stereotype threat. Educational Psychology Review,
25(2), 211–243. https://doi.org/10.1007/s10648-013-9219-1
Tibbetts, Y., Harackiewicz, J. M., Canning, E. A., Boston, J.
S., Priniski, S. J., & Hyde, J. S. (2016). Affirming indepen-
dence: Exploring mechanisms underlying a values affirma-
tion intervention for first-generation students. Journal of
Personality and Social Psychology, 110(5), 635–659. https://
doi.org/10.1037/pspa0000049
van Veelen, R., Derks, B., & Endedijk, M. D. (2019). Double trouble:
How being outnumbered and negatively stereotyped threatens
career outcomes of women in STEM. Frontiers in Psychology,
10, Article 150. https://doi.org/10.3389/fpsyg.2019.00150
Walton, G. M. (2014). The new science of wise psychological
interventions. Current Directions in Psychological Science,
23(1), 73–82. https://doi.org/10.1177/0963721413512856
Walton, G. M., Logel, C., Peach, J. M., Spencer, S. J., & Zanna, M.
P. (2015). Two brief interventions to mitigate a “chilly climate”
transform women’s experience, relationships, and achievement
in engineering. Journal of Educational Psychology, 107(2),
468–485. https://doi.org/10.1037/a0037461
Wright, M. C., McKay, T., Hershock, C., Miller, K., & Tritz, J.
(2014). Better than expected: Using learning analytics to
promote student success in gateway science. Change: The
Mechanisms Underlying Values Affirmation Interventions
19
Magazine of Higher Learning, 46(1), 28–34. https://doi.org/10.
1080/00091383.2014.867209
Yeager, D. S., & Walton, G. M. (2011). Social-psychological
interventions in education: They’re not magic. Review of
Educational Research, 81(2), 267–301. https://doi.org/10.3102/
0034654311405999
Zedelius, C. M., Mills, C., & Schooler, J. W. (2019). Beyond
subjective judgments: Predicting evaluations of creative writ-
ing from computational linguistic features. Behavior Research
Methods, 51(2), 879–894. https://doi.org/10.3758/s13428-018-
1137-1
Authors
NIA M. M. DOWELL is an assistant professor in the School of
Education at the University of California, Irvine. Dowell’s team
conducts basic research on sociocognitive and affective processes
across a range of educational technology interaction contexts and
develops computational models of these processes and their rela-
tionship to learner outcomes.
TIMOTHY A. MCKAY is an Arthur F. Thurnau Professor of
Physics, Astronomy, Education, and associate dean for
Undergraduate Education at the University of Michigan. His
research focuses on exploring grading patterns and performance
disparities both at Michigan and across the CIC, developing a vari-
ety of data-driven student support tools like E2Coach through the
Digital Innovation Greenhouse, an innovation space for exploring
the personalization of education, and launching the National
Science Foundation–funded REBUILD project.
GEORGE PERRETT is the director of Research and Data Analysis
at New York University. His work sits at the intersection of
machine learning and causal inference.
... They then look at 12,500 openended survey responses from these students, use latent Dirichlet allocation (LDA) to identify topics or thematic resources mentioned, and use those to guide qualitative coding of specific resources FIGs' afford. Similar to the Dowell et al. (2021) article, they use data science tools to discern why an experimental condition has its observed effects (e.g., the seminars offer integration, belonging and information). ...
... Whereas the Dowell et al. (2021) and Aulck et al. (2021) articles use NLP methods to understand how and why a treatment is effective or not, the article by Kylie Anglin, Vivian Wong, and Arielle Boguslav (Anglin et al., 2021; "A Natural Language Processing Approach to Measuring Treatment Adherence and Consistency Using Semantic Similarity") uses NLP to determine whether an educational reform is being delivered with fidelity. The work argues that the success of most reforms and interventions depends on whether it is actually implemented as proposed. ...
Article
Full-text available
This AERA Open special topic concerns the large emerging research area of education data science (EDS). In a narrow sense, EDS applies statistics and computational techniques to educational phenomena and questions. In a broader sense, it is an umbrella for a fleet of new computational techniques being used to identify new forms of data, measures, descriptives, predictions, and experiments in education. Not only are old research questions being analyzed in new ways but also new questions are emerging based on novel data and discoveries from EDS techniques. This overview defines the emerging field of education data science and discusses 12 articles that illustrate an AERA-angle on EDS. Our overview relates a variety of promises EDS poses for the field of education as well as the areas where EDS scholars could successfully focus going forward.
Technical Report
Full-text available
The words that people use in everyday life tell us about their psychological states: their beliefs, emotions, thinking habits, lived experiences, social relationships, and personalities. From the time of Freud’s writings about “slips of the tongue” to the early days of computer-based text analysis, researchers across the social sciences have amassed an extensive body of evidence showing that people’s words have tremendous psychological value. To appreciate some of the truly great pioneers, check out (Allport, 1942), Gottschalk and Gleser (1969), Stone et al., (1966), and Weintraub (1989). Although promising, the early computer methods floundered because of the sheer complexity of the task. In order to provide a better method for studying verbal and written speech samples, we originally developed a text analysis application called Linguistic Inquiry and Word Count, or LIWC (pronounced “Luke”). The first LIWC application was developed as part of an exploratory study of language and disclosure (Francis & Pennebaker, 1992). The second (LIWC2001), third (LIWC2007), fourth (2015), and now fifth (LIWC-22) versions updated the original application with increasingly expanded dictionaries and sophisticated software design (Pennebaker et al., 2001, 2007, 2015). The most recent evolution, LIWC-22 (Pennebaker et al., 2022), has significantly altered both the dictionary and the software options to reflect new directions in text analysis. As with previous versions, the program is designed to analyze individual or multiple language files quickly and efficiently. At the same time, the program attempts to be transparent and flexible in its operation, allowing the user to explore word use in multiple ways.
Article
Full-text available
Scholars across disciplines have long debated the existence of a common structure that underlies narratives. Using computer-based language analysis methods, several structural and psychological categories of language were measured across ~40,000 traditional narratives (e.g., novels and movie scripts) and ~20,000 nontraditional narratives (science reporting in newspaper articles, TED talks, and Supreme Court opinions). Across traditional narratives, a consistent underlying story structure emerged that revealed three primary processes: staging, plot progression, and cognitive tension. No evidence emerged to indicate that adherence to normative story structures was related to the popularity of the story. Last, analysis of fact-driven texts revealed structures that differed from story-based narratives.
Conference Paper
Full-text available
Women are traditionally underrepresented in science, technology, engineering, and mathematics (STEM). While the representation of women in STEM classrooms has grown rapidly in recent years, it remains pedagogically meaningful to understand whether their learning outcomes are achieved in different ways than male students. In this study, we explored this issue through the lens of language in the context of an asynchronous online discussion forum. We applied Linguistic Inquiry and Word Count (LIWC) to examine linguistic features of students’ reflective posting in an online chemistry class at a four-year university. Our results suggest that cognitive linguistic features significantly predict the likelihood of passing the course and increases perceived sense of belonging. However, these results only hold true for female students. Pronouns and words relevant to social presence correlate with passing the course in different directions, and this mixed relationship is more polarized among male students. Interestingly, the linguistic features per se do not differ significantly between genders. Overall, our findings provide a more nuanced account of the relationship between linguistic signals of social/cognitive presence and learning outcomes. We conclude with implications for pedagogical interventions and system design to inclusively support learner success in online STEM courses.
Chapter
Full-text available
Summarization is an effective strategy to promote and enhance learning and deep comprehension of texts. However, summarization is seldom implemented by teachers in classrooms because the manual evaluation of students’ summaries requires time and effort. This problem has led to the development of automated models of summarization quality. However, these models often rely on features derived from expert ratings of student summarizations of specific source texts and are therefore not generalizable to summarizations of new texts. Further, many of the models rely of proprietary tools that are not freely or publicly available, rendering replications difficult. In this study, we introduce an automated summarization evaluation (ASE) model that depends strictly on features of the source text or the summary, allowing for a purely text-based model of quality. This model effectively classifies summaries as either low or high quality with an accuracy above 80%. Importantly, the model was developed on a large number of source texts allowing for generalizability across texts. Further, the features used in this study are freely and publicly available affording replication.
Article
Full-text available
A wide range of occupations require science, technology, engineering, and mathematics (STEM) skills, yet almost half of students who intend to pursue a postsecondary STEM education abandon these plans before graduating from college. This attrition is especially pronounced among underrepresented groups (i.e., racial/ ethnic minorities and first-generation college students). We conducted a 2-year follow-up of a utility-value intervention that had been implemented in an introductory biology course. This intervention was previously shown to improve performance in the course, on average and especially among underrepresented students, reducing the achievement gap. The goal of the present study was to examine whether the intervention also impacted persistence in the biomedical track throughout college. The intervention had a more positive impact on long-term persistence for students who were more confident that they could succeed at the beginning of the course, and this effect was partially driven by the extent to which students reflected on the personal relevance of biological topics in their essays. This mechanism was distinct from the process that had been found to underlie intervention effects on performance-engagement with course material-suggesting that utility-value interventions may affect different academic outcomes by initiating distinct psychological processes. Although we did not find that the intervention was differentially effective for underrepresented students in terms of persistence, we found that positive effects on performance were associated with increased persistence for these students. Results suggest that utility-value interventions in an introductory course can be an effective strategy to promote persistence in the biomedical sciences throughout college.
Article
Large amounts of resources are spent annually to improve educational achievement and to close the gender gap in sciences with typically very modest effects. In 2010, a 15-min self-affirmation intervention showed a dramatic reduction in this gender gap. We reanalyzed the original data and found several critical problems. First, the self-affirmation hypothesis stated that women’s performance would improve. However, the data showed no improvement for women. There was an interaction effect between self-affirmation and gender caused by a negative effect on men’s performance. Second, the findings were based on covariate-adjusted interaction effects, which imply that self-affirmation reduced the gender gap only for the small sample of men and women who did not differ in the covariates. Third, specification-curve analyses with more than 1,500 possible specifications showed that less than one quarter yielded significant interaction effects and less than 3% showed significant improvements among women.
Article
Collaborative problem-solving (CPS) has become an essential component of today’s knowledge-based, innovation- centred economy and society. As such, communication and CPS are now considered critical 21st century skills and incorporated into educational practice, policy, and research. Despite general agreement that these are important skills, there is less agreement on how to capture sociocognitive processes automatically during team interactions to gain a better understanding of their relationship with CPS outcomes. The availability of naturally occurring educational discourse data within online CPS platforms presents a golden opportunity to advance understanding about online learner sociocognitive roles and ecologies. In this paper, we explore the relationship between emergent sociocognitive roles, collaborative problem-solving skills, and outcomes. Group Communication Analysis (GCA) — a computational linguistic framework for analyzing the sequential interactions of online team communication — was applied to a large CPS dataset in the domain of science (participant N = 967; team N = 480). The ETS Collaborative Science Assessment Prototype (ECSAP) was used to measure learners’ CPS skills, and CPS outcomes. Cluster analyses and linear mixed-effects modelling were used to detect learner roles, and assess the relationship between those roles on CPS skills and outcomes. Implications for future research and practice are discussed regarding sociocognitive roles and collaborative problem-solving skills.
Article
Collection and analysis of students' writing samples on a large scale is a part of the research agenda of the emerging writing analytics community that promises to deliver an unprecedented insight into characteristics of student writing. Yet with a large scale often comes variability of contexts in which the samples were produced-different institutions, different purposes of writing, different author demographics, to name just a few possible dimensions of variation. What are the implications of such variation for the ability of automated methods to create indices/features based on the writing samples that would be valid and meaningful? This paper presents a case study in system generalization. Building on a system developed to assess the expression of utility value (a social-psychology-based construct) in essays written by first-year biology students at one postsecondary institution, we vary data parameters and observe system performance. From the point of view of social psychology, all these variants represent the same underlying construct (i.e., utility value), and it is thus very tempting to think that an automatically produced utility-value score could provide a meaningful analytic, consistently, on a large collection of essays. However, findings from this research show that there are challenges: Some variations are easier to deal with than others, and some components of the automated system generalize better than others. The findings are then discussed both in the context of the case study and more generally.
Article
Utility-value (UV) interventions, in which students complete writing assignments about the personal usefulness of course material, show great promise for promoting interest and performance in introductory college science courses, as well as persistence in science, technology, engineering, and mathematics fields. As researchers move toward scaling up this intervention, it's important to understand which features are key to its effectiveness. For example, prior studies have used different types of UV assignments (i.e., self-focused essays and other-focused letters) and different assignment structures (i.e., over time, researchers provided a variety of tasks or choices between tasks), without comparing them. It is not known whether these assignment features are incidental details or key aspects of the intervention that impact its effectiveness. In the current study, we systematically compared different UV assignments, as well as ways of combining them, in a randomized controlled trial in an introductory college biology course (N = 590). Specifically, we compared different versions of the intervention in terms of their relative effectiveness for promoting course performance and the motivational mechanisms through which they operated. The intervention was most effective when students had opportunities to write about utility for both the self and others. Grades were higher in conditions in which students were either assigned a variety of self-focused and other-focused assignments or given the choice between the two. Among students with low performance expectations, grades were higher when students were assigned a specific combination: a self-focused assignment followed by other-focused assignments. Results suggest that different versions of the intervention may work through different mechanisms.