ArticlePDF Available

Identifying Second Language Speech Tasks and Ability Levels for Successful Nurse Oral Interaction with Patients in a Linguistic Minority Setting: An Instrument Development Project


Abstract and Figures

One of the most demanding situations for members of linguistic minorities is a conversation between a health professional and a patient, a situation that frequently arises for linguistic minority groups in North America, Europe, and elsewhere. The present study reports on the construction of an oral interaction scale for nurses serving linguistic minorities in their second language (L2). A mixed methods approach was used to identify and validate a set of speech activities relating to nurse interactions with patients and to derive the L2 ability required to carry out those tasks. The research included an extensive literature review, the development of an initial list of speech tasks, and validation of this list with a nurse focus group. The retained speech tasks were then developed into a questionnaire and administered to 133 Quebec nurses who assessed each speech task for difficulty in an L2 context. Results were submitted to Rasch analysis and calibrated with reference to the Canadian Language Benchmarks, and the constructs underlying the speech tasks were identified through exploratory and confirmatory factor analyses. Results showed that speech tasks dealing with emotional aspects of caregiving and conveying health-specific information were reported as being the most demanding in terms of L2 ability, and the most strongly associated with L2 ability required for nurse-patient interactions. Implications are discussed with respect to the development and use of assessment instruments to facilitate L2 workplace training for health care professionals.
Content may be subject to copyright.
Health Communication, 26: 560–570, 2011
Copyright © Taylor & Francis Group, LLC
ISSN: 1041-0236 print / 1532-7027 online
DOI: 10.1080/10410236.2011.558336
Identifying Second Language Speech Tasks and Ability Levels for
Successful Nurse Oral Interaction with Patients in a Linguistic
Minority Setting: An Instrument Development Project
Talia Isaacs
Graduate School of Education
University of Bristol
Michel D. Laurier
Department of Management and Foundations of Education
University of Montreal
Carolyn E. Turner
Department of Integrated Studies in Education
McGill University, and Centre for the Study of Learning and Performance
Norman Segalowitz
Department of Psychology
Concordia University, and Centre for the Study of Learning and Performance
One of the most demanding situations for members of linguistic minorities is a conversation
between a health professional and a patient, a situation that frequently arises for linguistic
minority groups in North America, Europe, and elsewhere. The present study reports on the
construction of an oral interaction scale for nurses serving linguistic minorities in their second
language (L2). A mixed methods approach was used to identify and validate a set of speech
activities relating to nurse interactions with patients and to derive the L2 ability required to
carry out those tasks. The research included an extensive literature review, the development of
an initial list of speech tasks, and validation of this list with a nurse focus group. The retained
speech tasks were then developed into a questionnaire and administered to 133 Quebec nurses
who assessed each speech task for difficulty in an L2 context. Results were submitted to
Rasch analysis and calibrated with reference to the Canadian Language Benchmarks, and the
constructs underlying the speech tasks were identified through exploratory and confirmatory
factor analyses. Results showed that speech tasks dealing with emotional aspects of caregiv-
ing and conveying health-specific information were reported as being the most demanding in
terms of L2 ability, and the most strongly associated with L2 ability required for nurse–patient
interactions. Implications are discussed with respect to the development and use of assessment
instruments to facilitate L2 workplace training for health care professionals.
There is growing interest in the nature of health com-
munication (Candlin & Candlin, 2003) and, in particular,
communication problems that can arise when patients and
caregivers do not speak the same first language (L1)
Correspondence should be addressed to Talia Isaacs, Graduate School
of Education, University of Bristol, 35 Berkeley Square, Bristol BS8 1JA,
UK. E-mail:
(Robinson, 2002). One concern about such language gaps is
that communication may suffer to the point of compromis-
ing health care quality. For example, a language barrier may
make it difficult to establish good caregiver–patient rapport
and, as a consequence, the patient may receive less atten-
tive care and experience less satisfaction with the medical
treatment than patients for whom there is no language bar-
rier (Carrasquillo, Orav, Brennon, & Burstin, 1999; Jacobs,
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
Chen, Karliner, Agger-Gupta, & Mutha, 2006). In some
areas of health care—including the delivery of bad news
(e.g., Gillotti, Thompson & McNeilis, 2002), obtaining
informed consent (e.g., Meyer, 2003; Schenker, Wang, Selig,
Ng, & Fernandez, 2007), or talking about death, suicide,
pain or other sensitive issues—good communication with
the patient can be especially critical for obvious emotional,
ethical, and legal reasons (Anderson et al., 2003; Chen,
Youdelman, & Brooks, 2007). More generally, t h e r e c a n b e
communication problems between patient and health profes-
sionals even with the same L1 (e.g., Cegala et al., 2008; Ong,
de Haes, Hoos, & Lammes, 1995), and these problems are
likely exacerbated when one speaker has a poor command
of the language used.
In linguistically diverse countries such as the United
States and Canada and especially in those with several offi-
cial language communities (e.g., Canada, India), providing
equitable health care services in multiple languages can
be a challenge (Bélanger, 2003; Bowen, 2001; Canadian
Medical Association, 2005; Lavizzo-Mourey, 2007). Often
this entails health professionals speaking in their nonfluent
second language (L2) or enlisting the aid of an interpreter
(frequently a medically unqualified family member or hos-
pital personnel; Fernandez & Schenker, 2010; Lee et al.,
2006). Such situations can also result in the patient accom-
modating linguistically to the caregiver rather than the other
way around. It can be expected, therefore, that increasingly
there will be occasions when caregivers do not feel they
have the necessary L2 competence to handle their patients’
communicative needs with the sensitivity and comprehen-
siveness they deserve.
Steps can be taken to address such language barriers
(Schenker, Lo, Ettinger, & Fernandez, 2008). One is to iden-
tify those situations that caregivers perceive to be difficult to
handle in an L2 and to target the relevant sociolinguistic and
pragmatic functions in L2 training. While more is becoming
known about the nature of communication between patients
and caregivers with the same L1 (e.g., Roter & McNeilis,
2003; see also papers in Heritage & Maynard, 2006a), rel-
atively little is known about such communication when the
health professional uses a nonfluent L2. Systematic research
on this topic is sorely needed. Having a tool capable of
identifying areas of communication that caregivers view as
presenting significant barriers in using their L2 would enable
researchers to focus on the most vulnerable aspects of med-
ical communication. This should aid the development of
interventions to overcome language barriers. The present
study investigated a way to identify areas of communica-
tion where such language barriers are likely to exist between
nurses and patients.
Before going into the study, however, it will be useful
to consider how language barriers might manifest them-
selves. One approach to this is provided by speech act
theory (Austen, 1962; Searle, 1969, 1976). The challenge
nurses face when using the L2 is not simply to translate
L1 words and expressions into their L2. Rather, they must
communicate intentions (perform illocutionary rather than
just locutionary or utterance acts). People carry out illocu-
tionary acts in a wide variety of ways, including by using
complex forms of indirect speech (e.g., “Do you think you
might be able to do something about that smoking habit?”).
Formal speech act analysis requires careful consideration of
the context in which communication takes place. Moreover,
as Blum-Kulka (1997) and Cooren (2005) have shown,
such formal analysis also requires taking into account the
sequencing of the conversational exchange. Thus, analyses
cannot be restricted to isolated speech samples examined
in the abstract. Because illocutionary acts are expressed in
conventionalized ways that vary across linguistic cultures
(Blum-Kulka, 1997), language barriers likely arise at this
level. It is important, therefore, to understand how speech
acts are normally expressed in the target language and to
identify the kinds of miscommunication and misunderstand-
ing that arise from failure to handle speech acts appropriately
in the L2.
Conversation analysis (Drew, 2005; Heritage & Maynard,
2006b) provides another perspective on language barriers.
Here the focus is the “negotiation” that characterizes com-
munication. If the caregiver has low L2 proficiency or is
unfamiliar with L2 conversational routines and conventions,
then the natural flow of the conversational negotiation may
be compromised, creating a language barrier. Yet another
approach to language barriers comes from an intercultural
pragmatics perspective (Wierzbicka, 2008). Here, culturally
specific semantic scripts are held to underlie the meanings
of certain words (e.g., for key English words like “personal”
and “privacy”). Failure to appreciate the semantic script
underlying key words can result in language barriers.
Thus, for any broadly defined speech activity (e.g., con-
veying bad news about a prognosis, or obtaining informed
consent) there are required speech acts, rules for conversa-
tional negotiation, and cultural scripts that must be handled
properly in the L2 to avoid linguistic barriers. Such inter-
cultural differences in language use have implications for
L2 training (Knutson, 2006) aimed at overcoming barriers.
An important first step, however, is to systematically identify
the communicative activities in which language barriers are
likely to arise. Once the potential locus of a language bar-
rier has been found, the underlying nature of the barrier can
be investigated, the results of which should inform remedial
To sum mar i ze, t he goa l of ov erc omin g lang uag e b arr iers
in health care settings minimally requires three steps—
identifying communicative tasks where significant language
barriers are likely to arise; closely examining those tasks
using various analytical tools; and developing appropriate
remedial steps. The present study focuses on the first step.
Rather than targeting communicative tasks a priori using
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
process, with the results subsequently submitted to quantita-
tive analyses. For the qualitative phase, domain experts (i.e.,
nurses) were consulted about their language use to guide the
selection of communicative activities to be examined. The
literature on nurse–patient communication has been grow-
ing (Shattell, 2003) and includes case studies documenting
what nurses actually say in particular situations (Skott &
Eriksson, 2005; Wadensten, 2005). No studies that we are
aware of, however, have systematically investigated nurses’
speech activities using an L2, nor their perceptions of task
relevance and difficulty. The present research aims to fill
this void. In a second phase, the data were analyzed quan-
titatively (Rasch analysis; factor analyses) to expand upon
and refine the qualitative results. The main emphasis of this
study is on this second, quantitative phase.
The motivation for this project arose from within the health
care community and the desire of health practitioners to
linguistically accommodate their patients. Direct communi-
cation with patients as opposed to reliance on third parties
is also in the researchers’ interest. The general target group
is in-service registered nurses who work in their L2 when
serving linguistic minorities in Quebec, Canada.
The goal of the study is threefold: (1) to identify and
validate a set of speech tasks that nurses regard as the
most relevant when serving patients in their nondominant
language; (2) to identify the dimensions underlying these
speech tasks, in order to better understand the L2 skills
needed to carry them out; and (3) to calibrate the nursing-
specific speech tasks onto a previously validated rating scale
commonly used for workplace language training, in order
to extend the applicability of the scale to nursing-specific
communicative tasks. This study is conceived of as a step-
ping stone to the eventual development of an oral assessment
instrument designed to enhance L2 teaching and learning of
health-specific oral content for nurses. The subject of this
article is the development and validation of a list of rele-
vant speech tasks, the first step in this longer-term research
agenda. The specific research questions are:
RQ1: To what extent do nurses identify, in a consistent
and reliable way, specific speech tasks that they per-
ceive as relevant and requiring an appropriate level
of language ability in L2 interactions with patients?
RQ2: To what extent can the speech tasks so identified
be scaled in relation to existing L2 performance
RQ3: To what extent do the patterns of speech tasks iden-
tified by nurses fall into well-defined categories that
could serve as the basis for developing a pedagog-
ically oriented assessment instrument for nurse L2
The study used a sequential exploratory mixed meth-
ods design commonly used for instrument development
(Creswell & Plano Clark, 2007). The qualitative and quan-
titative data were collected sequentially across two stages.
Decisions made at the first stage (qualitative) influenced the
direction of the second stage (quantitative). In addition, the
first stage results helped guide the second stage in instances
of quantitative analysis limitations. The qualitative methods
included a literature review, focus group, and verbal proto-
col. The product of this first stage was a questionnaire for
nurses. The quantitative methods included Rasch analysis
and factor analyses of the questionnaire data. The product of
the second stage was a list of speech tasks and ability levels
that are specific and relevant to the nursing profession.
Methodology and Results for the Qualitative Stage
As part of the questionnaire development, we wanted to
align the speech tasks with a well-known, previously vali-
dated oral interaction scale commonly used to assess adult
L2 speakers. The Canadian Language Benchmarks (CLB),
French, was chosen for this purpose (Pawlikowska-Smith,
2000). Proficiency levels are indicated as reference points
(benchmarks) on a 12-point scale grouped into three cate-
gories: Beginner, Intermediate and Advanced.
Preliminary Questionnaire
The qualitative stage began with an extensive literature
review that centered on identifying nursing-specific speech
tasks. The goal was to generate a list of speech tasks for
a preliminary questionnaire to be used with a focus group
of nurses. Twenty-two speech tasks were identified and a
preliminary questionnaire was developed with scales for
relevance and difficulty.
Five L1 French nurses representative of diverse care
domains in a Quebec hospital (Mexperience =27.8 years;
20–35) participated in a 2 hour focus group led by two of the
researchers. Using the preliminary questionnaire as a guide,
the principal goals were to identify relevant speech tasks
from the list provided; gather specific contextual examples;
estimate the linguistic difficulty necessary to perform these
speech tasks in their L2 (i.e., English); and add to the list
(e.g., nurses deemed the speech task “rephrasing patients’
ideas” to be relevant and “rather difficult” to perform in an
L2 due to its lexical demands). This process generated 19
speech tasks. Some were deleted due to irrelevancy, and new
relevant ones were added.
Verb al Pr o toc ol
In order to refine the list of speech tasks to be included
in the questionnaire, a verbal protocol was administered to
an experienced L1 French nurse. This was pursued as an
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
alternative to piloting for validation purposes. Using a think-
aloud procedure, the nurse was audio recorded verbalizing
her thoughts in French as she reflected on the language level
needed to perform each speech task in her L2 (English) using
the CLB scale. At the end of the process, she reflected, “If the
task is of a technical nature, then telling them [patients] what
they have to do is simple. But uh . . . it’s much more complex
when you have to deal with the patient’s emotions.” The data
were transcribed and subjected to a qualitative content anal-
ysis. The 19 speech tasks from the focus group were retained
with slight wording modification. The English version of the
speech tasks appears in the appendix.
Methodology and Results for the Quantitative Stage:
Up to this point (i.e., the first stage), we had identified rel-
evant speech activities for nurse–patient interaction. Next,
we developed a final questionnaire. The goal was to reach
out to a larger population of nurses to validate relevancy
and identify the proficiency level needed to perform each
speech task. In considering the CLB descriptors and content
analysis of the verbal protocol, it was evident that the lower
and higher levels included language functions less relevant
to nursing. We decided to align our proficiency levels for the
final questionnaire with the middle seven levels; thus, the
CLB levels 4 to 10 became our levels 1 to 7. Two formats of
a final questionnaire were developed to accommodate par-
ticipants: a hard copy bilingual questionnaire, and an online
questionnaire with two versions (French/English) developed
through the online questionnaire development tool Survey
Monkey ( Participants were
provided with the 19 speech tasks and the 7-level CLB
descriptors and were asked to identify the relevant CLB level
for each speech task using the prompt, “The person who can
perform the following task . . . should at least be at level.
The participants who completed the questionnaire were
159 registered nurses (reduced to 133 after further analyses;
see later discussion), who were familiar with the day-to-
day realities of the health care context and were, therefore,
deemed well positioned to make judgments about language
use in their profession (Mexperienc e =19.2 years; 1–35). They
were recruited from language schools, the Quebec nurse
licensing body, hospitals, a conference, and a local univer-
sity. The highest sampling came from the first two groups
(79 nurses from nine different language schools and 30 from
the licensing body). In addition, variation by geographical
region was achieved, with nurse respondents from 11 out of
the 17 regions in Quebec. From self-reported data we learned
of their language use at work. Since the majority were L1
French speakers (n=110), these data are reported here. On
ascaleof1to4(1=never;4=always), the mean French
use at work was 3.89 (SD =.31); the mean English use was
2.22 (SD =.58); and the mean use of another language was
1.20 (SD =.40). Finally, their mean L2 English proficiency,
as reported on a 3-point scale (1 =beginner,3=advanced),
was 1.71 (SD =.66).
Descriptive statistics were calculated using SPSS 14.0.
Rasch analysis was then employed using FACETS 3.60.0
to refine the data, determine the relevance of the 19 speech
tasks, and align them with the CLB scale for L2 proficiency
level. Finally, factor analyses were performed using SAS
9.1, in order to identify the underlying constructs being
The questionnaire data were analyzed for the purpose of
determining the degree of agreement among the nursing
community about the 19 speech tasks. From this perspective,
gence among the nurses. This convergence was observed for
each indicator in reference to the following criteria:
Correspondence between the level previously identi-
fied by the members of the focus group and the nurse
questionnaire respondents’ estimates of the necessary
level to perform a speech task.
Differences between respondents in their individual
estimates: When the nurses agree, they tend to assign
the same level to a given speech task.
Consistency of their ratings in relation with other indi-
cators: The speech task ranking in terms of difficulty
is similar from one respondent to the other.
In order to identify the indicators that show convergence,
we used a combination of descriptive statistics (mean, SD,
correlation) and Rasch analysis, a methodology used in pre-
vious studies (e.g., Blais, Laurier, & Rousseau, 2009). Using
Rasch analysis software, we applied a polychotomous Rasch
model to analyze the ordered categories obtained from the
rating scale (Andrich, 1999). Seven categories correspond-
ing to the seven proficiency levels were recognized using
a simple two-facet model (indicators ×respondents). This
type of analysis allows one to model the data instead of
just describing it. The difficulty indices are more robust than
the mean scores because they are less sample dependent. In
addition, indices about the fit of the indicators and respon-
dents’ (nurses’) ratings in relation to the model can yield
valuable information about the consistency of their response
Rasch analysis is iterative, and three iterations were
needed to clean up the data. This process resulted in 26
nurse respondents being removed from the original 159,
due to either invariant ratings across speech tasks or erratic
(unmodeled) rating behavior. The final sample size, there-
fore, was 133 nurses. Generally, one would expect a signif-
icant proportion of the items (indicators) to be rejected due
to poor model fit. Interestingly, all 19 speech tasks were val-
idated. That is, nurse respondents did agree on what should
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
and should not be considered difficult tasks. We mainly con-
sidered infit mean squares due to their utility for identifying
indicators that may threaten the validity of the measurement
system and ease of interpretation. This index is based on
the chi-squared statistic with each observation weighted by
its model variance. The expected value is 1. Values below
1 indicate overfit, which in this situation is unproblematic,
as some redundancy is expected. Values above 1 indicate
underfit because of noise or sources of variance that add
on the main variance, reflecting variations in relation with
the trait being measured. When the fit index is above 2,
we can conclude that the indicator or the respondent may
corrupt the measurement system. The analysis revealed that
all indicators had infit indices below Wright and Linacre’s
(1994) suggested cutoff value of 1.4. Thus, all items were
The process of calibration aims at locating each facet
element on the “Rasch ruler.” This common scale usually
ranges from 4to+4 logits. The analysis is normally set to
fix the 0 point at the mean item difficulty (in this case, the
speech task), and speech tasks and raters (in this case, nurse
respondents) are located accordingly. Because the Rasch
model uses logits as the measurement unit, the scale is con-
sidered arithmetic (e.g., an indicator calibrated at 2 is twice
as difficult as an indicator calibrated at 1). The placement
of respondents on the ruler is interpreted as the degree of
severity of their ratings.
In Figure 1, the first column (column R) represents the
“Rasch ruler.” Only the 2to+2rangeisshownbecause
few observations fall outside these limits. The column L on
the right shows where the boundaries between levels are
located. The most difficult speech tasks as judged by the
nurses appear at the top of the “items” column, while the
least difficult appear at the bottom. As predicted, the indica-
tor “Inform patient of bad news” is clearly the most difficult
task to perform by a nurse in an L2, whereas “Check identity
over phone” was the easiest overall. The other speech tasks
are grouped within the 1and+1range.Thefarrightcol-
umn evidences the wide distribution of the nurses (raters) in
relation to severity.
Ideally an indicator should be neither too difficult nor
too easy. For a given speech task, all questionnaire respon-
dents should identify the same difficulty level. In practice,
this situation never occurs because raters never reach perfect
consensus. When the respondents tend to agree on a level,
the distribution around the mean is not skewed, the mode is
obvious and the standard deviation is low. Figure 2 shows
such a distribution for the indicator “Making encouraging
statements to a patient.” The mean (3.79), the Rasch mea-
sure (0.89), and the mode (4) confirm the nurses’ proposal to
consider that the level required to perform this speech task
is level 4 (level 7 of the CLB). The low standard deviation
(1.49) can be interpreted as general agreement among the
nursing community, and this hypothesis is sustained by the
infit mean square, which is close to 1.
In summary, the Rasch analysis confirmed that all speech
tasks were functioning adequately from a statistical stand-
point (i.e., without undue distortion) and should be retained
in the subsequent analyses. At the same time, the nurses
whose fit indices were not predicted by the model—that is,
whose responses were either too erratic or unvaried to be
modeled—were removed from the data set. This served to
“clean up” the data so that patterns could be more easily
discerned in the subsequent analyses.
The questionnaire data were subjected to an exploratory
factor analysis using the varimax rotation. Multiple indices
suggested that three factors should be extracted. First, the
first eigenvalue was preeminent, accounting for 73.5% of
the variance; however, the specks in the scree plot did not
level off until the fourth eigenvalue. This suggests that the
first three eigenvalues accounted for a substantial proportion
of the variance (93.6% cumulatively). Second, the applica-
tion of Kaiser’s criterion, which holds that only eigenvalues
greater than one should be retained, led to the selection of
three eigenvalues (Stevens, 2002). Finally, the overall root
mean square residual was less than .05, which implies that
the reproduced matrix with three factors represented the
original data well.
Following the selection of three factors preliminarily,
the data were subjected to a confirmatory factor analysis.
Items with high residuals are usually discarded in this pro-
cedure in the interests of attaining better model fit and for
reasons of parsimony. However, because the Rasch item
indices and descriptive statistics had provided evidence that
all 19 speech tasks were functioning adequately for the pur-
pose of developing a nursing-specific descriptive scale, it
was decided a priori that no speech tasks would be dis-
carded in the factor analyses. Furthermore, the focus-group
nurses had deemed these speech tasks to be the most rele-
vant to the nurse–patient interactive tasks. This substantive
argument overrode any statistical expedience of excluding
speech tasks. Several models were assessed under the con-
straint of retaining all speech tasks, and the second-order
confirmatory factor analysis model shown in Figure 3 was
found to be the best. One higher order factor (F4) predicted
three second-order factors (F1, F2, and F3), all of which, in
turn, predicted the 19 speech tasks.
the validity of the model. Schermelleh-Engel, Moosbrugger,
and Müller’s “rule of thumb” ranges of good model fit were
adopted as a guideline (2003). In fact, none of the indices
approached the good fit range. The pvalue for the χ2test,
for example, was <.0001 (good fit range: .05 p.10),
the RMSEA estimate was .11 (good fit range: 0 RMSEA
.05), CFI was.89 (good fit range: .97 CFI 1.0), and
NNFI was .85 (good fit range: .97 NNFI 1.0). These
fit indices would clearly not be acceptable for high-stakes
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
19) Inform patient of bad news
----------------------------------------- ---------------------------------------
13) Clarify doctor's message
18) Manage patient's anger
10) Reason with patient
7) Summarize health situation
11) Reassure patient
6) Give info about services
12) Show empathy
8) Describe reaction to diagnosis
5) Give instructions on phone
4) Give instructions in person
9) Describe interventions
16) Apologize to patient
15) Refuse unreasonable requests
17) Encourage patient
14) Ask routine questions
3) Reformulate pain description
2) Give directions over phone
1) Check identity over phone
FIGURE 1 Location of the indicators on the “Rasch Ruler.”
Note. Only nonextreme scalar responses (scale levels 2–6) are shown here.
test development (e.g., a nurse licensing exam). However, the
result was deemed sufficient for the overall goal of deriving
a descriptive rating instrument designated for pedagogical
use, provided that the speech task clustering was readily
interpretable in the process of labeling the factors.1
The speech tasks with the highest factor loadings (shown
in parentheses) for factor one (F1) were: managing a
patient’s anger or impatience (.94); informing a patient of
bad news (.81); refusing unreasonable requests made by a
1As outlined earlier, letting the numbers drive the data by “throwing
out” speech tasks was not pursued in this study. This was due to the nature
of the low-stakes instrument being developed and to the importance placed
on the domain experts’ input in the development and validation of the list
of speech tasks. However, in the interests of examining whether the data
could support a better fitting model if statistically redundant speech tasks
were to be removed, models that optimized fit by excluding speech tasks
with large unique variance components were examined. A three-factor
solution with nine speech tasks yielded the best fitting model, and all
indices fell within the Schermelleh-Engel et al. (2003) “good fit” range
(p=.22; RMSEA =.04; CFI =.99; NNFI =.99). This suggests that an
assessment instrument for nurse licensure purposes could potentially be
created using the same data in light of fit statistics; however, this was not
the purpose of the present study.
patient (.80); making apologies to a patient (.79); showing
empathy towards a patient (.79); and reassuring a patient
(.70). The descriptions “anger,” “apologies,” “empathy,”
“reassuring,” and managing a patient’s likely adverse reac-
tion to receiving bad news and potentially negative reaction
to having an unreasonable request refused suggest a fac-
tor that deals with managing a patient’s emotions. F1 was
therefore labeled Emotional aspects of caregiving.
The speech tasks with the highest factor loadings for
factor two (F2) were: giving instructions to a patient
accurately on the phone (1.02) or face to face (.99);
rephrasing/confirming a patient’s description of pain (.80);
summarizing a patient’s health situation (.80); and providing
information about available services in relation to a patient’s
medical condition (.75). Giving instructions or informa-
tion and summarizing, paraphrasing, or acknowledging a
patient’s description of pain seemed to delineate a factor that
entails conveying information about a patient’s medical con-
dition. F2 was therefore labeled Fac tua l a spe cts rel ate d to
the patient’s medical condition.
The speech tasks with the highest factor loadings for fac-
tor three (F3) were: checking a patient’s identity over the
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
Item 17: Making encouraging
statements to a patient (e.g.,
patient beginning to walk
Rasch Item Indices
Difficulty: –0.89
Infit (MnSq): 1.07
Sample Statistics:
n = 131 (out of 133)
M = 3.79
SD = 1.49
10 2345
Scale levels
FIGURE 2 Indicator with an adequate distribution.
Task # Factor loadings
2nd order
1st order
FIGURE 3 Second-order confirmatory factor analysis model. While the arrows show all of the relationships specified in the structural equations for
the confirmatory factor analysis model, only statistically significant factor loadings are shown in the factor loading boxes (α=.01, two-tailed). The
number appearing in the second line of these boxes after the “#” sign refers to the speech task item number that the factor is linked with. Boxes that
contain “D” are the error terms associated with the second-order factors; boxes that contain “E” terms are the error terms associated with the speech
tasks. Notably, there is no correlation between the second-order factors (i.e., they are orthogonal).
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
phone (1.02); giving instructions on how to get to the hos-
pital over the phone (.94); refusing unreasonable requests
made by a patient (.85); and making encouraging statements
to a patient (.84). “Checking a patient’s identity,” giv-
ing directions over the phone, and offering encouragement
seemed to relate to communicating with a patient on non-
health-specific topics. Notably, the speech task “refusing
unreasonable requests” also loaded onto F1, the emotional
factor. This double loading can perhaps be explained by
the scenario where refusing a patient’s unreasonable request
could potentially escalate into needing to deal with his/her
emotions. Conversely, refusing a patient’s request to smoke
on the premises, the example that was provided with the
speech task (see the appendix), might be considered quite
routine or uninvolved. F3 was therefore labeled Routine
Finally, the factor loadings between F4 and the second-
order factors were as follows. The loading onto F1,
Emotional aspects of caregiving,was.76.Theloadingonto
F2, Fac t ual a spe cts rel ate d to th e pat ien t ’s m edi cal c ond i-
tion, was .55. Finally, the loading onto F3, Routine aspects,
was .23. Both F1 and F2 were significantly associated
with F4 at the p<.01 level (Stevens, 2002). Thus, factual
aspects and emotionally charged aspects of caregiving, both
of which are health specific, relate to the overarching con-
struct. F3, however, was not significantly correlated with
F4. In fact, the ability represented by F3 appears to run
contrary to the development of F4, in that the correlation
was weak and negative. This may be because the formu-
laic language involved in carrying out speech tasks such
as “checking a patient’s identity” or “making encouraging
statements” has little to do with the L2 ability required to
respond to emotionally-sensitive health topics or convey spe-
cific information about a patient’s health. F4 was therefore
labeled Health-specific verbal interaction with patients.
Having completed the Rasch and factor analyses, what
remained was to align the 19 speech tasks with the CLB scale
bands so that these nurse–patient interactive tasks could be
directly associated with a CLB level. First, each speech task
was linked with the second order factor with the highest
factor loading (F1, F2, or F3). Next, Rasch item indices and
descriptive statistics were examined to assign each speech
task to the appropriate CLB level. Although the reduced
CLB scale with seven levels was used in the questionnaire
administered to nurses, Figure 4 shows that the indicators
were only calibrated onto five CLB scale bands. That is, the
resulting scale was a 5-point scale. Notably, the easiest items
were all associated with F3, the routine factor. Due to their
generic (i.e., non-health-specific) nature, these speech tasks
may not be instructive for getting at health-specific nurse–
patient communication, the construct of interest (F4). The
most salient speech tasks in this regard (i.e., those most
CLB band level Speech tasks
# 1
# 2 #14 #17
# 3 #15 #16 #9 #4
#12 # 5 # 6 #11 #7 #10 #18
# 8 #13 #19
F1 =
F2 =
F3 =
More difficult item
FIGURE 4 Aligning speech tasks with CLB scale bands.
strongly associated with F1 or F2), were aligned with CLB
scale bands 7–9.
Summary and Contributions
The purpose of this study was to develop an oral interac-
tion scale for nurses serving linguistic minorities in their L2.
Qualitative findings informed the development and imple-
mentation of a quantitative instrument in a sequential man-
ner. In the qualitative stage, an extensive literature review
on oral interactions between health practitioners and patients
and input from focus group nurses led to a draft list of the
most pertinent speech activities. This list was developed into
and distributed to nurses working in different contexts across
Quebec. In the quantitative stage, the questionnaire data
were analyzed through descriptive statistics and Rasch mod-
eling and adequate item functioning was confirmed. Next,
the factor analyses, with all 19 speech tasks included as
exogenous variables, resulted in grouping the indicators into
three dimensions, which, in turn, were underpinned by the
broader dimension of health-specific oral interaction skills
for nurses (F4). Speech tasks dealing with a patient’s emo-
tions (F1) and conveying health-specific information (F2)
were the most pertinent. The speech tasks related to routine
aspects of nurses’ professional life (F3), on the other hand,
bore little relation to the construct of interest.
That emotionally charged speech tasks are generally
more difficult to carry out in an L2 than routine tasks may
not surprise either domain experts or “lay” people outside
the health care system. Indeed, people often make intuitive
judgments about the relative difficulty of the tasks they carry
out in their daily lives and extrapolate from experience. A
main contribution of the study is that it provides an empir-
ical basis for determining which speech tasks are the most
difficult and relevant. While nurses are not linguists, they
were arguably the best placed to gauge the emotional impact
of an interactive task and how that might render a particular
speech task easier or more difficult to carry out in an L2, a
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
perspective that should be taken into account in the L2 train-
ing and testing of nurses. The perspectives of other health
professionals (e.g., doctors, pharmacists) could similarly be
probed using a mixed methodology.
Notably, the speech tasks were calibrated onto the CLB
band descriptors. This national standard has exerted an influ-
ence on Canadian classrooms that train adult L2 learners for
professional purposes, including newcomers seeking to inte-
grate into the health care sector. Aligning the speech tasks
with the CLB bands extends the applicability of the scale to
health care professionals. In addition, it provides informa-
tion on the CLB level necessary for nurses to be able to carry
out interactive tasks in their professional domain (CLB lev-
els 7–9 for the health-specific speech tasks in this study).
This complements other attempts to devise assessment tools
for Canadian nurses using the CLB like the CELBAN,which
was created for licensure purposes (CELBAN, 2010).
The descriptive scale developed in this study is part of
to help health practitioners improve their L2 skills when
dealing with minority language patients in the workplace.
Conceived of as a tool for teaching and learning, this instru-
ment will include assessment tasks that center on emotional
aspects of caregiving and factual aspects related to patients’
medical conditions. The challenge will be to design an
instrument that can be easily integrated into typical class-
room and self-training activities. The speech tasks could be
expanded into a series of role plays, for example, which
have been a mainstay for the training of health care profes-
sionals (e.g., Walker, Cedergren, Trofimovich, Gatbonton,
& Mikhail, 2008; Watt & Lake, 2007). The skill devel-
opment resulting from such an instrument could enhance
nurses’ willingness to engage in using their L2 in situations
they might otherwise shy away from and, thus, better serve
minority communities.
The interplay between qualitative and quantitative meth-
ods is an advantage of the present study. However, several
limitations need to be taken into account in interpreting
the results. First, the focus group was relatively small; a
more extensive focus group consultation may have yielded a
larger set of candidate communicative activities to examine.
However, Creswell and Plano Clark (2007) emphasize that
in sequential exploratory mixed-methods designs, a small
sample size is appropriate in the initial qualitative explo-
ration. Second, the test sample was limited to in-service
nurses from different regions of Quebec. Results are, to
some extent, reflective of the Quebec health care needs and
system. The study would need to be replicated to deter-
mine how well the findings generalize. Obtaining data from
This is particularly the case for the factor analyses, since
the study fell short of the norm of 10 participants for each
factor. However, the use of qualitative data arguably com-
pensated for this shortcoming. Third, the study was limited
to nurses’ perspectives of their language needs. A necessary
next step is to conduct a needs analysis to gauge the per-
spectives of other health professionals charged with different
daily communicative tasks (e.g., doctors), with the long-
term goal of enhancing workplace-specific L2 training. The
“Essential Skills” inventory developed by Human Resources
and Skills Development Canada may be a useful starting
point in this regard (de Vries, 2009). Finally, the study
investigated nurses’ language needs when serving patients
in their L2 with a focus on English or French, Canada’s offi-
cial languages. The choice of these languages understates
the multilingual reality of Quebec society (Lamarre, 2003),
however, and ignores contexts in which health practitioner
training in nonofficial languages may be beneficial (e.g.,
Inuit and First Nations languages). Future research could
focus on instrument development to enhance nurse training
in additional languages.
This study reported a mixed methods approach for system-
atically identifying health care-relevant speech tasks and
placing them in an ordered list. The ranking of these nurse–
patient interactive tasks reflects both nurses’ perceptions
of the linguistic difficulty of the task, and nurses’ self-
assessments of their ability to effectively communicate in
their L2 in that task situation. The resulting list of speech
tasks can inform fine-grained follow-up analyses of lin-
guistic barriers aimed at supporting the development of L2
training for nurses. In such follow-up analyses, it must be
remembered that skill in using the L2 is not just a matter of
knowing how to translate from L1 to L2. As Heritage and
Maynard (2006b) point out, “It is by acting together that
doctor [caregiver] and patient assemble each particular visit
with its interactional textures, perceived features and out-
comes. . . . [The] ordinary norms and practices of language
use and social interaction exert a powerful and systematic
influence on the texture and features of medical visits, and do
so in fine detail” (pp. 19–20). As pointed out earlier, how a
communicative event unfolds when the caregiver is speaking
in the L2 is interesting from a variety of linguistic stand-
points (speech act theory, conversation analysis, intercultural
pragmatics, etc.). Understanding how caregivers perceive the
challenges inherent in participating in these events is a cru-
cial first step in the process of developing appropriate L2
training for caregivers.
This research was made possible by support to the Health-
Care Access for Linguistic Minorities (H-CALM) research
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
team, which is part of the Training and Human Resources
Development Project (THRDP) based at McGill University,
funded by Health Canada. The research was also supported
in part by a grant from the Social Sciences and Humanities
Research Council of Canada to NS and CT. Some of
the data from this study were previously presented at the
annual meeting of the Canadian Association of Applied
Linguistics (2009, Ottawa, Canada) and at the Language
Testing Research Colloquium (2010, Cambridge, UK). We
are grateful to Maia Yarymowich, our research coordinator,
for her support throughout the project, to the many institu-
tions and individuals who helped us get access to nurses, and
to our nurse participants.
Anderson, L., Scrimshaw, S., Fullilove, M., Fielding, J., & Norman, J.
(2003). Culturally competent healthcare systems: A systematic review.
American Journal of Preventative Medicine,24, 68–79.
Andrich, D. (1999). Rating scale analysis. In J. P. Keeves & G. N. Masters
(Eds.), Advances in measurement in educational research and assessment
(pp. 110–121). Oxford: Pergamon.
Austin, J. L. (1962). How to do things with words. Cambridge, MA: Harvard
University Press.
Bélanger, M. (2003). L’accès aux soins de santé pour les communautés
minoritaires de langue officielle. Ottawa: Chambres des Communes,
Blais, J.-G., Laurier, M. D., & Rousseau, C. (2009). Deriving profi-
ciency scales from performance indicators using the Rasch model. In
E. V. Smith, Jr., & G. E. Stone (Eds.), Criterion referenced testing:
Practice analysis to score reporting using Rasch measurement models
(pp. 528–540). Maple Grove, MN: JAM Press.
Blum-Kulka, S. (1997). Discourse pragmatics. In T. A. van Dijk (Ed.),
Discourse as social interaction, Volume 2 (pp. 38–63). London: Sage.
Bowen, S. (2001). Language barriers in access to health care.Ottawa:
Health Canada.
Burnard, P. (2003). Ordinary chat and therapeutic conversation: Phatic com-
munication and mental health nursing. Journal of Psychiatri c and Mental
Health Nursing,10, 678–682.
Canadian Medical Association. (2005). National physician survey:New
data reflect multiculturalism’s impact on medicine. Retrieved from
Candlin, C., & Candlin, S. (2003). Health care communication: A prob-
lematic site for applied linguistics research. Annual Review of Applied
Linguistics,23, 134–154.
Carrasquillo, O., Orav, J., Brennon, T., & Burstin, H. (1999). Impact of
language barriers on patient satisfaction in an emergency department.
Journal of General Internal Medicine,14, 82–87.
Cegala, D. J., Bahnson, R. R., Clinton, S. K., David, P., Gong, M. C., Monk,
J. P. III . . . Pohar, K. S. (2008). Information seeking and satisfaction
with physician–patient communication among prostate cancer survivors.
Health Communication,23, 62–69.
CELBAN. (2010). Canadian English Language Benchmark Assessment
for Nurses. Retrieved from
Chen, A. H., Youdelman, M. K., & Brooks, J. (2007). The legal framework
for language access in healthcare settings: Title VI and beyond. Journal
of General Internal Medicine,22, 362–367.
Cooren, F. (2005). The contribution of speech acct theory to the analy-
sis of conversation: How pre-sequences work. In K. L. Fitch & R. E.
Sanders (Eds.), Handbook of language and social interaction, (pp.
21–40). Mahwah, NJ: Lawrence Erlbaum Associates.
Creswell, J. W., & Plano Clark, V. L. (2007). Designing and conducting
mixed methods research. Thousand Oaks, CA: Sage.
de Vries, E. (2009). Workplace essential skills in practice: A Canadian per-
spective. In R. Maclean & D. Wilson (Eds.), International handbook of
education for the changing world of work (pp. 2713–2730). Dordrecht,
The Netherlands: Springer.
Drew, P. (2005). Conversational analysis. In K. L. Fitch & R. E. Sanders
(Eds.), Handbook of language and social interaction (pp. 71–102).
Mahwah, NJ: Lawrence Erlbaum Associates.
Fernandez, A., & Schenker, Y. (2010). Time to establish national stan-
dards and certification for health care interpreters. Pat ien t Ed uca ti on a nd
Counseling 78, 139–140.
Gillotti, C., Thompson, T., & McNeilis, K. (2002). Communicative com-
petence in the delivery of bad news. Social Science & Medicine,54,
Heritage, J., & Maynard, D. W. (Eds.). (2006a). Communication in
medial care: Interaction between primary care physicians and patients.
Cambridge: Cambridge University Press.
Heritage, J., & Maynard, D. W. (2006b). Introduction: Analysing inter-
action between doctors and patients in primary care encounters. In J.
Heritage & D. W. Maynard (Eds.), Communication in medial care:
Interaction between primary care physicians and patients (pp. 1–21).
Cambridge: Cambridge University Press.
Jacobs, E., Chen, A., Karliner, L. S., Agger-Gupta, N., & Mutha, S. (2006).
The need for more research on language barriers in health care: A
proposed research agenda. Milbank Quarterly,84, 111–133.
Knutson, E. M. (2006). Cross-cultural awareness for second/foreign lan-
guage learners. Canadian Modern Language Review,62, 591–610.
Lamarre, P. (2003). Growing up trilingual in Montreal: Perceptions of
college students. In R. Bayley & S. R. Schecter (Eds.), Language social-
ization in bilingual and multilingual societies (pp. 62–80). Clevedon,
UK: Multilingual Matters.
Lavizzo-Mourey, R. (2007). Improving quality of US health care hinges on
improving language services. Journa l of General Internal Medicine,22,
Lee, K. C., Winickoff, J. P., Kim, M. K., Campbell, E. G., Betancourt, J.R.,
Park, E. R. . . .Weissman, J. S. (2006). Resident physicians’ use of pro-
fessional and nonprofessional interpreters: A national survey. Journal of
the American Medical Association,296, 1050–1053.
Meyer, B. (2003). Bilingual risk communication. Arbeiten zur
Mehrsprachigkeit, 50. Universität Hamburg.
Ong, L., de Haes, J., Hoos, A., & Lammes, F. (1995). Doctor–patient com-
munication: A review of the literature. Social Science and Medicine,40,
Pawlikowska-Smith, G. (2000). Canadian Language Benchmarks 2000.
Ottawa, ON: Centre for Canadian Language Benchmarks.
Robinson, M. (2002). Communication and health in a multi-ethnic society.
Bristol: Policy Press.
Roter, D., & McNeilis, K.S. (2003). The nature of the therapeutic relation-
ship and the assessment and consequences of its discourse in routine
medical visits. In T. Thompson, A. Dorsey, K. Miller, & R. Parrott
(Eds.), Handbook of health communication (pp. 121–140). Mahwah, NJ:
Lawrence Erlbaum Associates.
Schenker, Y., Lo, B., Ettinger, J. D., & Fernandez, A. (2008). Navigating
language barriers under difficult circumstances. Annals of Internal
Medicine,149, 264–269.
Schenker, Y., Wang, F., Selig, S. J., Ng, R., & Fernandez, A. (2007). The
impact of language barriers on documentation of informed consent at
ahospitalwithon-siteinterpreterservices.Journal of Ge neral In ternal
Medicine,22, 294–299.
Schermelleh-Engel, K., Moosbrugger, H., & Müller, H. (2003). Evaluating
the fit of structural equation models: Tests of significance and descriptive
goodness-of-fit measures. Methods of Psychological Research Online,8,
Searle, J. (1969). Speech acts. Cambridge, UK: Cambridge University
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
Searle, J. (1976). A classification of illocutionary acts. Language in Society,
5, 1–23.
Shattell, M. (2003). Nurse–patient interaction: A review of the literature.
Journal of Clin ical Nursing,13, 714–722.
Skott, C., & Eriksson, A. (2005). Clinical caring: The diary of a nurse.
Journal of Clin ical Nursing,14, 916–921.
Stevens, J. P. (2002). Applied multivariate statistics for the social sciences
(4th ed.). Mahwah, NJ: Lawrence Erlbaum Associates.
Wadensten, B. (2005). The content of morning time conversations between
nursing home staff and residents. Journal of Clini cal Nurs ing,14,8489.
Wal ke r , N. , Ce dergr en , H. J . , Trofi mo vi ch, P., G atb on ton , E. , & Mi kha il , E.
(2008). Someone to talk to: A virtual patient for medical history inter-
view training in a second language. INTED (International Technology,
Education, and Development Conference) Proceedings, Valencia, Spain.
Wat t, D. , & La ke , D. (2 00 7) . Medical communication assessment project
(M-CAP): An Alberta project for international medical graduates. Report
prepared for the M-CAP, Faculty of Medicine, University of Calgary,
Calgary, Alberta, Canada.
Wierzbicka, A. (2008). A conceptual basis for intercultural pragmatics
and world-wide understanding. In M. Pütz & J. Neff-van Aertselaer
(Eds.), Developing contrastive pragmatics (pp. 3–45). Berlin: Mouton
de Gruyter.
Wright, B. D., & Linacre, J. M. (1994). Reasonable mean-square fit values.
Rasch Measurement Transactions,8, 370. Retrieved from http://www.
1. Checking a patient’s identity over the phone.
2. Giving instructions on how to get to the hospital or
clinic over the phone.
3. Rephrasing/confirming a patient’s description of
4. Giving instructions to a patient accurately face to
face (e.g., regarding medication).
5. Giving instructions to a patient accurately on the
phone (e.g., regarding medication).
6. Providing additional information about available ser-
vices in relation to the patient’s condition (e.g., after
7. Summarizing a patient’s health situation to the
8. Summarizing/rephrasing a patient’s feelings in reac-
tion to a diagnosis.
9. Describing to a patient in detail the common profes-
sional interventions (e.g., blood tests, injections).
10. Reasoning with a patient (e.g., after surgery has been
11. Reassuring a patient (e.g., in an emergency situation,
before surgery).
12. Showing empathy towards a patient.
13. Clarifying to a patient what the doctor tried to
explain regarding his/her medical condition.
14. Asking routine questions to a patient.
15. Refusing unreasonable requests made by a patient
(e.g., can’t smoke).
16. Making apologies to a patient (e.g., right before an
intervention that might be painful/uncomfortable,
forgot to do something requested by patient).
17. Making encouraging statements to a patient (e.g.,
when patient beginning to walk again).
18. Managing a patient’s anger or impatience.
19. Informing a patient of bad news (e.g., condition has
become worse).
Downloaded by [Concordia University Libraries] at 13:18 23 September 2011
... However, little attention is paid to the inspection of interaction processes of healthcare providers in foreign languages. With the advent of globalisation, English for medical purposes has been steadily increasing all over the world (Master 2005), and it has gradually become an international language in the field of public healthcare, too. Medical practitioners in Latvia, where free movement of the citizens of the European Union (EU) has its place in conformity with one of its fundamental principles, are not an exception, which means that local healthcare providers are also required to perform different professional activities in English if a situation requires it (Standard 2003, 21;Standard 2011, 3). ...
... Atslēgvārdi: saziņas kompetence, profesionālā darbība, angļu valoda, medicīnas speciālists, gadījuma izpēte. (2017) and Isaacs et al. (2011).) ...
Full-text available
The enhancement of Interactional Competence (IC) in English is a topical issue in the sphere of Public Health in many European countries, including Latvia, due to worldwide globalisation. IC competence of healthcare specialists is expected to be developed so that they can communicate with foreign patients requiring medical assistance in English. This is the awareness of the situations in which the English language is required most often that determines the effectiveness of medical practitioners’ IC enhancement. Therefore, the article aims to elucidate the most widespread professional oral activities healthcare specialists, such as nurses and physician assistants, are presumed to perform in English providing foreign patients with medical care or treatment. In order to reach the aim, a case study, including such data collection tool as a questionnaire, was used as a research method. The results of the research revealed that nurses and physician assistants most often encounter the situations when they need to obtain and provide data concerning a patient and his/her health condition as well as give a patient information about his/her further care or treatment in English. Keywords: interactional competence, professional activity, English language, healthcare specialist, case study. Reference: Pašinska, O. (2022) Identifying Professional Activities for Healthcare Specialists’ Oral Interactional Competence Enhancement in EFL. Language Aquisition: Problems and Perspectives, XVII/XVIII: 204-225. Data base: EBSCO and Crossref.
... It has to be highlighted that the choice of the topic and subtopics is regulated by the laws issued by the Cabinet of the Ministers (2019a, 2019b) and the Ministry of Education and Science (2004Science ( , 2020 concerning the healthcare specialists' competences and skills, as well as previous findings (Burrus & Willis, 2017;Isaacs et al., 2011). It implies that different materials and activities were developed to teach students to perform such tasks during classes. ...
Full-text available
About the use of various modalities in ESP classes applying digital platforms and tools Reference: Pašinska, O and Selivanova, J. (2022) Recommendations for the Use of Digital Tools and Learning Materials in Multimodal ESP Classrooms: a Case Study In K. Molek-Kozakowska (eds.) Multimodality of Academic Genres in Multilingual Education (pp. 58-101). ERASMUS+: Council of Europe ISBN: 978-83-7395-958-3
... Moving to the discussion of the capabilities of a non-native speaker to communicate in a foreign language, it should be admitted that poor foreign language knowledge is an obstacle to physician-patient communication that may result in hindering effective interaction (Isaacs et al., 2011, Travaline et al., 2005. If a specialist has a low level of foreign language proficiency as well as is unaware of the situational, social and cultural settings of a working place, the flow of conversation can get distorted (Isaac et al., 2011). ...
Full-text available
Healthcare specialists are expected to possess competences essential for a successful fulfilment of their duties. Although specialists’ practical competences enabling them to deal with a wide range of health conditions are the key ones, communicative competence, implying an ability to use a target language “accurately, appropriately, and flexibly” (Yule, 2010, p.194), has also had an immense impact on the accomplishment of professional duties (World Health Organisation, 2005). It follows that apart from giving opportunities to acquire knowledge about medicine and practical skills, medical education institutions are expected to teach 21st century students to communicate effectively using various linguistic tools applicable in their professional context (Zethsen and Askehave, 2006). The present article aims to elucidate the peculiarities of communicative competence that practising or future healthcare providers should possess in order to demonstrate how it enables them to accomplish their responsibilities by communicating with patients and, thus, provide them with qualitative care or treatment in a foreign language, specifically English as a foreign language. With the aim to reach the goal, a scientific literature review was conducted. The theoretical implications were supported by the examples of medical encounters; in the given encounters, the use of the selected language aspects and participants’ non-verbal cues were interpreted.
... 'can', 'might', 'should'), can increase readers' processing load compared to content words with only a single sense. The use of polysemous words compounds the challenge of extracting the correct meaning (Laufer, 1997), particularly for patients accessing information in their nondominant language, potentially contributing to communication difficulties (Isaacs et al., 2011). These examples illustrate that text difficulty cannot be adequately characterized using readability alone, which underrepresents the processing involved in reading text (McNamara et al., 2014). ...
Full-text available
Obtaining informed consent (IC) is an ethical imperative, signifying participants' understanding of the conditions and implications of research participation. One setting where the stakes for understanding are high is randomized controlled trials (RCTs), which test the effectiveness and safety of medical interventions. However, the use of legalese and medicalese in ethical forms coupled with the need to explain RCT-related concepts (e.g. randomization) can increase patients' cognitive load when reading text. There is a need to systematically examine the language demands of IC documents, including whether the processes intended to safeguard patients by providing clear information might do the opposite through complex, inaccessible language. Therefore, the goal of this study is to build an open-access corpus of patient information sheets (PIS) and consent forms (CF) and analyze each genre using an interdisciplinary approach to capture multidimensional measures of language quality beyond traditional readability measures. A search of publicly-available online IC documents for UK-based cancer RCTs (2000-17) yielded corpora of 27 PIS and 23 CF. Textual analysis using the computational tool, Coh-Metrix, revealed different linguistic dimensions relating to the complexity of IC documents, particularly low word concreteness for PIS and low referential and deep cohesion for CF, although both had high narrativity. Key part-of-speech analyses using Wmatrix corpus software revealed a contrast between the overrepresentation of the pronoun 'you' plus modal verbs in PIS and 'I' in CF, exposing the contradiction inherent Article 2 Health 00(0) in conveying uncertainty to patients using tentative language in PIS while making them affirm certainty in their understanding in CF.
... Given the requirement for many minority Francophones to converse with Anglophone health professionals, Isaacs, Laurier, Turner and Segalowitz (2011) tested an oral interaction scale for nurses working with linguistic minorities in their second language. Language tasks addressing the emotional aspects of healthcare delivery, as well as those involving the transmission of specific health information, were found to be the most demanding. ...
... Although there is an increased inclusion of elective global health experiences in nursing programs (29), the role of the nurses in these international experiences has received little attention (24). There is also a growing interest in health communication (30), especially when patients and caregivers do not share the same native language (31), but few studies have investigated communication among nurses and their perception of language barriers in their care (32). Moreover, no studies have investigated if there are differences between language abilities (bilingual or multilingual) of nurses and global health skills. ...
Full-text available
Objective Literature has reported that language is the most common barrier in a health care setting and a risk factor associated with negative outcomes. The present study reports the differences between nursing students who speak one language and nursing students who speak two or more languages (self-reported language proficiency) and their skills and learning needs in global health. Method An observational cross-sectional study was performed among nursing students from five Ontario universities. A survey was designed to measure knowledge, skills and learning needs in global health. Results Students who speak more than two languages are more likely to have more interest in learning global health issues, such as health risks and their association with travel and migration (p=0.44), and social determinants of health (p=0.042). Conclusion Language training is needed for nursing students to be able to face language barriers in health care settings and improve global health, locally and internationally.
[Full-text link:] This chapter pays tribute to Professor Cyril Weir’s major research into the nature of academic reading. Using one of his test development projects as an example, the chapter describes how the construct of academic reading was operationalised in the context of a British university by theoretical construct definition together with empirical analyses of students’ reading patterns on the test through eye-tracking. The chapter also reflects on how Weir’s various research projects fed into the development of the test and a new method of analysing eye-tracking data in relation to different types of reading.
Les auteurs se demandent si l’utilisation de marqueurs discursifs change, dans les interactions des infirmières avec les patients, en fonction de l’usage qu’elles font de leur langue maternelle (L1) ou de leur langue seconde (L2). Les marqueurs sont analysés à titre de marqueurs de maintien du tour de parole, indiquant la « prise en compte » et de marqueurs de transition dans le discours, indiquant un changement de sujet ou de locuteur dans la conversation. Ces deux catégories de marqueurs diffèrent sur le plan du degré de gestion du discours et du contrôle interactionnel. Seize infirmières mènent auprès d’un patient de L1 anglaise et d’un patient de L1 française un entretien visant l’évaluation de la douleur, entretien dans lequel les infirmières utilisent leur propre L1 dans un cas et leur propre L2, moins bien maîtrisée, dans l’autre. Les résultats de l’étude ne permettent pas de confirmer la première hypothèse, selon laquelle les infirmières devraient normalement utiliser les marqueurs discursifs plus fréquemment en L1 qu’en L2. Ils confirment toutefois la seconde hypothèse, selon laquelle les infirmières devraient utiliser les marqueurs de changement de sujet moins fréquemment en L2 qu’en L1, comparativement à leur usage (de base) des marqueurs de maintien du tour de parole. Les constatations des auteurs, en particulier l’attestation de la seconde hypothèse, pourraient avoir des répercussions sur le développement de la formation en L2 des professionnels de la santé.
This entry describes a study which falls into the category of instrument development for assessment and evaluation purposes using a mixed methods research (MMR) design.
Since the early 1990s, mixed methods research (MMR) has evolved into the third research paradigm alongside quantitatively oriented inquiry (numeric data) and qualitatively oriented inquiry (narrative data). MMR is interested in both quantitative and qualitative analyses and primarily works from a pragmatic stance to use what works best in order to answer research questions. The research question, rather than a preconceived paradigm (e.g., post-positivist, positivist, constructivist), is central and drives the choice of design. The interest in and practice of MMR have emerged rapidly and spread across many domains in the social and behavioral sciences. Within this context, the focus of this chapter is on discussing MMR as it has manifested itself within the language testing/assessment (LT) research community. In some ways MMR's evolution in LT parallels its general emergence, but in other ways there are developments specific to the LT field. For example, MMR is increasingly being employed for instrument development, classroom-based assessment, and large-scale assessment studies. In addition, MMR designs are being used to address issues such as construct definition and rater effects in language assessments. To help explain this evolution, the chapter situates MMR from an historical perspective and discusses how LT research designs are increasingly situating themselves in this third research community. As this trajectory is traced in LT research, the chapter also describes the nature, conventions, practice, and research designs that are emerging and becoming specific to the LT field. It is interesting, however, that any debate or discussion of MMR issues seems less prevalent in LT, whereas it is abundant in the general MMR literature. What is important to note is that LT's rationale for the use of MMR follows closely the philosophical orientation most often associated with it, that is, pragmatism.
Full-text available
In 1976, Patrick Byrne and Barrie Long published a path-breaking study of the doctor-patient relationship. Based on some 2,500 taperecorded primary care encounters, Doctors Talking to Patients anatomized the medical visit into a series of stages, and developed an elaborate characterization of doctor behaviors in each of them. Drawing on Michael Balint’s (1957) proposal that the primary care visit has therapeutic value in its own right, Byrne and Long focused on the ways in which its therapeutic possibilities were attenuated by the prevalence of doctor-centered behaviors in the encounters they studied. The study was also conceived as an intervention: physicians were invited to use its coding framework to evaluate their own conduct, and to modify it in a more patient-centered direction. Not surprisingly, given these goals, Doctors Talking to Patients was itself somewhat doctor-centered. The authors had little to say about patients’ contributions to the encounter or the sociocultural context of social interaction in primary care. In the present volume we revisit Byrne and Long’s project of anatomizing the primary care visit, doing so from a primarily sociological and interactional perspective.We begin from the standpoint that physician and patient-with various levels of mutual understanding, conflict, cooperation, authority, and subordination-jointly construct the medical visit as a real-time interactional product. Within this orientation, we consider some of the social, moral, and technical dilemmas that physicians and patients face in primary care, and the resources that they deploy in solving them.
This book provides a rigorous and challenging review of recent research in the realms of communication and cultural diversity. Focusing on health communication interventions concerning service users who may lack fluency in English, it shows that meeting the needs of all health service users depends on both structures and processes of communication.
Communication can be seen as the main ingredient in medical care. In reviewing doctor-patient communication, the following topics are addressed: (1) different purposes of medical communication; (2) analysis of doctor-patient communication; (3) specific communicative behaviors; (4) the influence of communicative behaviors on patient outcomes; and (5) concluding remarks. Three different purposes of communication are identified, namely: (a) creating a good inter-personal relationship; (b) exchanging information; and (c) making treatment-related decisions. Communication during medical encounters can be analyzed by using different interaction analysis systems (IAS). These systems differ with regard to their clinical relevance, observational strategy, reliability/validity and channels of communicative behavior. Several communicative behaviors that occur in consultations are discussed: instrumental (cure oriented) vs affective (care oriented) behavior, verbal vs non-verbal behavior, privacy behavior, high vs low controlling behavior, and medical vs everyday language vocabularies. Consequences of specific physician behaviors on certain patient outcomes, namely: satisfaction, compliance/adherence to treatment, recall and understanding of information, and health status/psychiatric morbidity are described. Finally, a framework relating background, process and outcome variables is presented.
This article is at