ArticlePDF Available

Abstract and Figures

As meta-analyses demonstrate feedback effects on performance, our study examined possible mediators. Based on our cognitive–motivational model [Vollmeyer, R., & Rheinberg, F. (1998). Motivationale Einflüsse auf Erwerb und Anwendung von Wissen in einem computersimulierten System [Motivational influences on the acquisition and application of knowledge in a simulated system]. Zeitschrift für Pädagogische Psychologie, 12, 11–23] we examined how feedback changed (1) strategies, and (2) motivation during learning, and by doing so improved (3) final performance. Students (N = 211) learned how a dynamic system works and how to reach given goal states for the system. One group received feedback (i.e., knowledge of performance) the other one did not. We expected learners to improve after they received the first feedback. However, we found that learners expecting feedback used better strategies right from the start. Thus, they acquired more knowledge over fewer trials. Although we had also expected effects of feedback on motivation during learning, we could not support this hypothesis.
Content may be subject to copyright.
A surprising effect of feedback on learning
Regina Vollmeyer
a,
*, Falko Rheinberg
b
a
Universita¨t Frankfurt, Germany
b
Universita¨t Potsdam, Germany
Abstract
As meta-analyses demonstrate feedback effects on performance, our study examined possible media-
tors. Based on our cognitiveemotivational model [Vollmeyer, R., & Rheinberg, F. (1998). Motivationale
Einflu
¨sse auf Erwerb und Anwendung von Wissen in einem computersimulierten System [Motivational
influences on the acquisition and application of knowledge in a simulated system]. Zeitschrift fu¨r
Pa¨dagogische Psychologie, 12,11e23] we examined how feedback changed (1) strategies, and (2) moti-
vation during learning, and by doing so improved (3) final performance. Students (NZ211) learned how
a dynamic system works and how to reach given goal states for the system. One group received feedback
(i.e., knowledge of performance) the other one did not. We expected learners to improve after they
received the first feedback. However, we found that learners expecting feedback used better strategies right
from the start. Thus, they acquired more knowledge over fewer trials. Although we had also expected
effects of feedback on motivation during learning, we could not support this hypothesis.
Ó2005 Elsevier Ltd. All rights reserved.
Keywords: Feedback; Motivation; Performance; Strategies
Educators have long regarded feedback as an important variable influencing learning.
Learners receiving feedback are provided with information about their performance, which
may tell them how well they have done a task and how to improve. However, the effects of
feedback are quite complex. These effects depend on: (a) the specific quality of feedback
learners receive; (b) the learners’ goals and motives (e.g., Nolen, 1996); and also (c) the
* Corresponding author. Institut fu
¨rPa
¨dagogische Psychologie, Johann Wolfgang Goethe-Universita
¨t Frankfurt,
Postfach 11 19 32, 60054 Frankfurt, Germany. Tel.: C49 331 977 2854; fax: C49 331 977 2791.
E-mail address: r.vollmeyer@paed.psych.uni-frankfurt.de (R. Vollmeyer).
0959-4752/$ - see front matter Ó2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.learninstruc.2005.08.001
Learning and Instruction 15 (2005) 589e602
www.elsevier.com/locate/learninstruc
learning context, particularly the consequences of good or poor performance. Feedback may
have an impact on different variables, for example on cognitive (attention, strategies, etc.),
motivational, and affective processes, as well as having enduring effects on the recipient’s
self concept (self-esteem, control, etc.).
As feedback is a core-concept in the theory and practice of teaching, it is not surprising that
it was one of the first variables experimental psychologists studied. Nearly 100 years ago
Thorndike (1913) started his work on the law of effect.Ammons’ (1956) summary of the
empirical research concluded that feedback (defined as knowledge of performance) generally
increased learning and motivation. Kluger and DeNisi (1996) presented an extensive historical
overview of literature regarding the effects of feedback interventions on performance, however,
they also found studies demonstrating that feedback could decrease performance. This finding
led Kluger and DeNisi to construct their Feedback Intervention Theory to try to integrate the
existing theories by including both motivational and cognitive effects of feedback. To combine
the motivational and cognitive effects of feedback, but in a more constrained way, we modified
our own cognitiveemotivational process model (Vollmeyer & Rheinberg, 1998) to take into
account feedback. Although our model was not specifically constructed to explain effects of
feedback, it provides a framework for studying which mediating variables influence the impact
of feedback on performance. Studies on mediating variables for the effect of feedback on
performance are not often reported in literature (Jussim, Soffin, Brown, Ley et al., 1992;
Schmidt & Kleinbeck, 1990).
1. Feedback in the cognitiveemotivational process model
Originally, our model was designed to describe the effects of cognitive and motivational
processes during learning. We assumed that initial motivation (i.e., four factors: challenge,
interest, anxiety, probability of success) helped learning via mediating variables, especially
strategy systematicity and motivational state during learning.Strategy systematicity indicates
how systematically learners explore the material, motivational state monitors aspects of
their motivation during learning such as how much fun learners have while learning, as well
as their confidence in increasing their knowledge (see also probability of success: Atkinson,
1957, 1964; Lewin, Dembo, Festinger, & Sears, 1944; and newer theories: Anderson, 1993;
self-efficacy: Bandura, 1986). Previous empirical studies (Vollmeyer & Rheinberg, 1999,
2000) showed that learners who had a favourable initial motivation used more systematic
strategies, and had a more positive motivational state (i.e., had more fun and were more
confident during learning). Higher levels of each of these mediator variables led to better learn-
ing outcomes.
As motivational state and strategy systematicity are relevant mediators for learning, we
assumed that a feedback manipulation should affect these variables. Feedback should lead to
higher motivation during learning and to a more effective, but more effortful strategy compared
to a condition without feedback. Thus with an experimental feedback manipulation we replaced
the function previously played by initial motivation (measured through a questionnaire) in our
model. However, our model still emphasises strategy systematicity and motivational state as
important mediating variables for effects on learning.
How could feedback influence these two mediators? According to Kulhavy and Wager’s
(1993) historical overview of feedback, during the first half of the 20th century feedback
was viewed in three ways: (1) as a motivator or incentive for increasing performance, (2) as
590 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
information which learners could use to validate or change previous response; and (3) as a sat-
isfying state of affairs. These functions of feedback still apply to the present day. Feedback as
a motivator is based on the belief that letting people know how well they are performing a task
will act as an incentive for greater effort in the future. Together with its informative function,
learners should think more about the structure of the task. Thus they should come up with
a more effective, but more effortful strategy compared to a condition without feedback. The
third function is clearly a part of what we understand as motivational state (e.g., having
more fun during learning). Therefore, we assumed that the feedback manipulation should affect
motivational state positively.
1.1. Hypotheses
In accordance with our cognitiveemotivational process model and empirical findings we de-
rived five hypotheses. Receiving feedback increases (1) the subsequent use of systematic
strategies; and increases (2) positive motivation (motivational state). This led to the next
hypothesis, that (3) knowledge acquisition should differ after receiving feedback: the feedback
group should gain more knowledge than the no feedback group. As the feedback group gained
more knowledge, (4) they should perform better when they had to apply their knowledge.
Finally (5) we expect that the two mediators, strategy systematicity and motivational state,
would mediate the effect of feedback on final performance.
2. Method
2.1. Participants
Two hundred and eleven university students in psychology (age 19e24) and high school
students (age 17e19) in Potsdam, Germany, participated in the study, for which they received
V12.00. From these participants 105 were told to receive feedback during learning, 106 did not
get this information.
2.2. The learning task biology lab
For studying feedback effects we needed a task for which we could precisely control the
conditions. However, in order to not get too far away from everyday life, the task should require
complex learning. Therefore learners should be free to choose their strategy and the amount of
time they wanted to spend. Our task simulated the everyday situation in which a learner
explored a complex system on a computer in a self-regulated way. Another advantage of this
task is that the learning process is well studied (Burns & Vollmeyer, 2002; Niemivirta, 2002;
Rheinberg, Vollmeyer, & Burns, 2002; Vollmeyer, Burns, & Holyoak, 1996; Vollmeyer &
Rheinberg, 1998, 1999, 2000).
The biology-lab task is a linear dynamic system (Funke, 1991) for studying problem solving.
In the instructions we gave participants the cover story that they were in a lab in which they
should test the effects of three medicines (A, B, C) on three substances found in the body
(Thyroxine, Histamine, Serotonin). The structure of the system (see Fig. 1) was such that
one output was influenced by only one input (Medicine A /Serotonin). One output (Thyrox-
ine) was affected by two inputs, and the other (Histamine) was affected by a decay factor
591R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
(marked as a circle connected to the output) in addition to a single input variable. The decay
factor was implemented by subtracting a percentage of the output’s previous value on each
manipulation of the system. Decay was a dynamic aspect of the system, because it yielded state
changes even if there was no manipulation (i.e., all inputs were set to zero).
There were two phases in which learners interacted with the task: in the learning phase, par-
ticipants had to acquire knowledge about how the input variables are connected to the output
variables. This phase consisted of three rounds, and in each round learners could manipulate
the inputs six times. Each manipulation is called a trial. After each trial learners immediately
saw the exact numbers of how much each body substance increased or decreased. This is a kind
of feedback too, however, learners did not get any information on the correctness of their
hypotheses they had about the system. In the application phase participants could apply their
knowledge in trying to reach goal states for each output variable. For each phase, we calculated
a performance measure. For the learning phase, we calculated how much participants learnt
about the system’s links, and for the application phase how close they got to the goal states
(goal achievement). Performance in this final application phase represents the level of
procedural knowledge (Anderson, 1993) the participants have acquired during the learning
phase.
To study the effects of feedback we told one group of participants that they will receive
feedback by the experimenter after the first and second round in the learning phase. However,
they would only receive feedback on how many links between the input and output variables
they correctly identified, they would be told nothing about the direction of the links (C/ÿ)
or their weights (e.g., ‘‘In this round you found three out of four existing relations.’’). This
kind of feedback manipulation is different to traditional feedback studies (see reviews by
Bangert-Drowns, Kulik, Kulik, & Morgan, 1991; Kulhavy & Stock, 1989; Kulik & Kulik,
1988) in which feedback is often manipulated as amount or type of information. We provided
minimal feedback in order to reduce the possibility that feedback could lead to better
performance simply because it provided information. Telling them their results for the links
only provided knowledge that learners usually acquired by the end of first round anyway as
demonstrated in Vollmeyer et al. (1996). We found that nearly all participants learned after
the first round which input variable affected which output variable. However, it was more
difficult for them to learn about the directions and weights. A second group served as a control
group that established how well participants learned the task if no feedback was given or
anticipated.
-.5
+4
-10
+6
Medicine A +2 Thyroxine
Histamine
Medicine B
Serotonin
Medicine C
Fig. 1. Structure of the biology-lab system.
592 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
2.3. Feedback manipulation
At the end of the instructions, both groups (feedback vs. no feedback) received the informa-
tion: ‘‘In total there are four connections between the three medicines and the substances found
in the body.’’ In addition, the feedback group was told: ‘‘After the first and second round the
experimenter will tell you how many of these four connections you found and also whether
you found the output variable with the decay factor.’
2.4. Measures
2.4.1. Mediating variables
Two mediating variables were measured in order to investigate the process of how feedback
influences learning: (1) strategy systematicity, and (2) motivational state.
2.4.1.1. Strategy systematicity. To measure how systematically participants explored the
system, we categorised each of the participants’ 18 opportunities to manipulate the system
into one of the following three categories. High systematicity was assigned when one input
was varied, and the other two were kept constant. For example, when a participant entered
10 for Medicine A, 0 for Medicine B, 0 for Medicine C. This strategy allowed participants
to discover the system’s structure, as they could observe which outputs changed in response
to the manipulation of a single input variable. Medium systematicity was assigned when
multiple inputs were varied, but a systematicity could be perceived. For example, a participant
entered 100 for Medicine A, 10 for Medicine B, 10 for Medicine C; or, 100 for Medicine A,
100 for Medicine B, ÿ100 for Medicine C. Low systematicity was assigned when a useful
systematicity could not be recognised. For example, a participant varied each input by the
same amount (e.g., 10 for Medicine A, 10 for Medicine B, 10 for Medicine C). To check
interrater reliability, 180 trials were coded by two raters and we received a Cohen’s
kZ0.94. All other trials were coded by only one rater.
To calculate a strategy systematicity score, we scored a trial’s systematicity from one to
three, with a value of three indicating a high systematic strategy for a trial, and one for
a low systematic strategy. Then we averaged the six trials per round to derive a systematicity
value for each round.
2.4.1.2. Motivational state. At the end of every learning round, participants answered three
questions on a seven-point scale that measured probability of success and positive valence
(‘‘The task is fun’’, ‘‘I’m sure I will find the correct solution’’, ‘‘It’s clear to me how to
continue’’). These items are homogenous (Cronbach’s aO0.83) (Round 1: MZ4.67,
SD Z1.36, Round 2: MZ4.39, SD Z1.32, Round 3: MZ4.67, SD Z1.36). We
decided to measure motivation during learning with only three items because we were afraid
that a longer questionnaire could interrupt the learning process. A long interruption could be
especially detrimental to motivation, the phenomenon we wanted to study.
At this point we need to make clear that this measure is still under development. Theoret-
ically we wanted to measure the same motivational factors as we had for initial motivation,
that is challenge, interest, probability of success and anxiety. However, measuring four
scales three times could be boring for the learners and thus it could impede their motivation.
Therefore we decided to use the term motivational state and to use only three items. It can
also be criticised as overlapping with metacognition, especially the item ‘‘It’s clear to me
593R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
how to continue’’. However, Weinert and Kluwe (1987) argued that this criticism is inherent to
the definition of these concepts. Given that we have not validated our construct yet, we will be
conservative and use the rather vague term motivational state.
2.4.2. Dependent variables
We measured performance for the learning phase (structure score) as well as for the appli-
cation phase (goal achievement).
2.4.2.1. Structure score. After each round of the learning phase, participants had to complete
a diagram, similar to the one in Fig. 1 but with all links and weights omitted. Using this
diagram they had to indicate their knowledge about the system’s structure by drawing a link
between an input and an output, if they noticed a relationship. Each link could be supplied
with a direction (Cor ÿ) and a weight, if participants thought they knew how strong the impact
was. To indicate that an output had a decay participants could write a weight into the empty
circle attached to an output.
The structure score consisted of (1) the number of correct links between the inputs and
the outputs, (2) the number of correct directions (C/ÿ), and (3) the number of correct weights.
The sums for the correct links and the correct directions were corrected for guessing (see
Woodworth & Schlosberg, 1954, p. 700). This structure score varied between 3.0 (best value)
and a theoretical minimum of ÿ1.8 (It was negative if participants incorrectly guessed too
much.) (Round 1: MZ1.36, SD Z1.07, Round 2: MZ1.58, SD Z1.07, Round 3:
MZ1.46, SD Z1.03).
2.4.2.2. Goal achievement. The final performance, that is, goal achievement in reaching the
goal state during the application phase, was computed as the sum of the absolute differences
between the target and the obtained number for each of the three output variables. As this
measure produced a skewed distribution, the variance was corrected by applying a logarithmic
transformation (ln). Goal achievement was computed for each of the six trials that comprised
each round in the application phase, in order to determine how close participants were able to
approach to the target goal. As there was no difference in performance between trials, the mean
error for the six trials was used. This difference score would mean that high scores were
indicators of poor performance. So that all performance measures would be in the same direc-
tion, all these scores were subtracted from an arbitrary constant. As said before, this indicates
level of procedural knowledge.
2.4.3. Procedure
Students received two pages of instructions about the biology lab. In the instructions the
feedback manipulation was administered. Then they started manipulating the inputs of the sys-
tem to try to induce the underlying structure shown in Fig. 1. After each of the three learning
rounds they filled out the structure diagram and answered the motivational state questionnaire.
The feedback group received feedback about how many links they correctly identified after
Rounds 1 and 2. Learners who spontaneously asked during the learning phase whether they
could receive the goal states earlier were allowed to start with the application round, but
they were not told that this opportunity was available. When the learning phase was finished,
participants were asked to reach the goal states, which were Thyroxine on 700, Histamine on
900, and Serotonin on 50. The experiment took about one and a half hours.
594 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
3. Results
When running the experiment we made an unexpected observation: there were more
participants who spontaneously stopped the learning phase and asked to receive the goal states
after the first or second rounds of the learning phase than we had found before (e.g., Vollmeyer
& Rheinberg, 1999, 2000). In our instructions we never said explicitly that there is an oppor-
tunity to end the learning phase earlier, therefore it was the participants’ wish to end learning.
Our procedure allowed them to terminate the learning phase if they wished to, which avoided
the possibility that participants’ motivation could be reduced by being forced to continue
an activity they regarded as useless. From a methodologist’s point of view the learners’ early
stoppage was undesirable, as it meant that all learners were not comparable. However, this
situation comes closer to reality, as this is what happens when people learn in a self-regulated
way. Therefore we decided to treat the round during which participants stopped as a dependent
variable and analyse what affected it.
Because of this surprising observation, and without having an a priori hypothesis, we
checked whether this behaviour was due to our feedback manipulation. Table 1 shows in which
round (Rounds 1e3) of each condition (feedback vs. no feedback) participants finished the
learning phase. It seemed that more participants in the feedback condition finished the learning
in the early rounds than in the no feedback condition. To check this conclusion we compared
how many participants in the feedback and no feedback conditions finished after Rounds 1 or 2
against how many waited until the end of Round 3, and found a significant difference,
c
2
(1) Z7.18, pZ0.007. As a consequence of this result we report statistical analyses testing
our hypotheses for the effects of feedback separately for each round because as the number of
rounds progressed the participants still doing the task may not be a random sample of those who
started the task. For example, learners finishing in an early round may have been more
motivated, may have gained enough knowledge, or they may have used more time for Round
1. After testing our original hypotheses, we performed explorative analysis examining possible
biases affecting which round participants finished in (see below).
3.1. Effects of feedback on the learning process
3.1.1. Hypothesis 1
Table 2 shows the differences between the feedback and the no feedback groups on the
mediating and dependent variables. (The strategy measure for three participants could not be
calculated because the records of their actions during the learning phase were incomplete.) It
appears that from the very beginning of their attempts to manipulate the system the groups
differed in that the feedback group chose more often the systematic strategy to change one
variable at a time, t(206) Z2.74, pZ0.007. In the second round, from the remaining partic-
ipants it was again the feedback group who varied the system more systematically,
Table 1
Number of participants finishing in each round of the learning phase per condition (feedback vs. no feedback)
End at.Feedback No feedback
Round 1 18 8
Round 2 26 18
Round 3 61 80
595R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
t(182) Z2.79, pZ0.006. In Round 3, after 44 participants had finished the learning phase, the
same effect of feedback on strategy systematicity was observed: participants in the feedback
group manipulated the system more systematically than the no feedback group,
t(137) Z2.95, pZ0.004. Thus although the group size diminished, the expected effect of feed-
back on strategy systematicity was repeatedly found.
3.1.2. Hypothesis 2
The second mediating variable, motivational state, was measured with three items after
each learning round. We checked whether our feedback manipulation affected motivation
during learning. Although the means of the feedback group for all three rounds were higher
than those for the no feedback group (means are depicted in Table 2) these differences were
not significant, Round 1: t(209) Z1.00, pZ0.32, Round 2: t(183) Z0.13, pZ0.90, Round
3: t(138) Z1.21, pZ0.23.
3.1.3. Hypothesis 3
We had proposed that the feedback group would learn more about the structure of the system
than the no feedback group. However, we had not expected there to be any impact of this
manipulation on knowledge until after Round 2, as participants did not receive their first
feedback until they had completed the structure diagram at the end of Round 1. Contradicting
this expectation are the knowledge scores reported in Table 2, which show that at the end of
each of the three rounds the feedback group had acquired more knowledge than the no feedback
group (Round 1: t(209) Z2.85, pZ0.005, Round 2: t(183) Z3.80, p!0.001, Round 3:
t(138) Z4.13, p!0.001).
3.1.4. Hypothesis 4
Knowing more about the structure of the system should help participants reach the goal
states, thus we expected the feedback group to come closer to the goal states. We confirmed
this hypothesis in finding that the feedback group came closer to the goal states (MZ4.70,
SD Z1.81) than the no feedback group (MZ4.15, SD Z1.85), t(206) Z2.17, pZ0.031.
Table 2
Descriptive statistics (M, SD, n) for how the feedback manipulation affected the learning phase (Rounds 1e3)
Feedback No feedback
Strategy systematicity (maximum Z3.0)
Round 1 2.39 (0.61, 103) 2.14 (0.67, 105)
Round 2 2.57 (0.51, 86) 2.33 (0.64, 98)
Round 3 2.64 (0.51, 60) 2.32 (0.69, 79)
Motivational state (maximum Z7.0)
Round 1 4.91 (1.43, 105) 4.71 (1.54, 106)
Round 2 4.67 (1.67, 87) 4.64 (1.60, 98)
Round 3 4.61 (1.59, 61) 4.29 (1.57, 79)
Structure score (maximum Z3.0)
Round 1 1.53 (1.09, 105) 1.10 (1.09, 106)
Round 2 1.87 (1.00, 87) 1.29 (1.09, 98)
Round 3 1.83 (0.95, 61) 1.13 (1.02, 79)
596 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
3.1.5. Hypothesis 5
According to our cognitiveemotivational model we expected strategy systematicity and
motivational state to mediate the effect of feedback on goal achievement. As Baron and Kenny
(1986) suggested, a variable (e.g., strategy systematicity) functions as a mediator when it meets
three conditions. In our case the three conditions to be fulfilled are (a) feedback should
influence strategy systematicity, (b) strategy systematicity should affect goal achievement
and (c) when feedback and strategy systematicity are controlled, the previously statistically
significant relation between feedback and goal achievement should no longer be significant.
We tested these three conditions for motivational state and strategy systematicity. To take
into account the fact that our design had three rounds and participants were finishing in
different rounds we analysed the values for motivational state and strategy systematicity for
each individual’s last round. For example, for learners finishing Round 2 we analysed the
motivational state and the strategy systematicity in Round 2. First we analysed whether moti-
vational state was a mediator. However, Baron and Kenny’s first condition could not be
supported, as feedback did not influence motivational state, bZ0.10, tZ1.50, pZ0.14.
Therefore motivational state could not be regarded as mediating the effect of feedback on
goal achievement although motivational state influenced goal achievement, bZ0.46,
tZ7.21, p!0.001.
The same analysis was run for strategy systematicity. Conditions 1 and 2 were fulfilled
in that feedback had an effect on strategy systematicity, bZ0.30, tZ4.36, p!0.001, and
strategy systematicity affected goal achievement (see Table 3). In a regression analysis (see
Table 3) with goal achievement as criterion and feedback and strategy systematicity as predic-
tors, feedback was no longer significant, but strategy systematicity still had an effect. Therefore
strategy systematicity can be regarded as a mediator.
In addition, to test the two mediators in one model we ran a path analysis. The correlations
between feedback, motivational state, strategy systematicity and goal achievement are pre-
sented in Tab l e 4. As a first model, we assumed that feedback correlates with both mediators e
strategy systematicity and motivational state eand those affect goal achievement. From the
correlations (Table 4) and previous studies (e.g., Vollmeyer & Rheinberg, 2000) we knew
that the mediators are correlated. Therefore we let the error terms correlate. This model had
a good fit, AGFI Z0.96, CFI Z0.99, RMSEA Z0.062, c
2
(1) Z1.74, pZ0.18, however,
the path from strategy systematicity to goal achievement turned out to be not significant. To
take more strongly into account that the mediators are correlated we analysed a model that
can be seen in Fig. 2. We now argue that feedback affects strategy systematicity, which then
Table 3
Summary of hierarchical regression analysis for variables predicting goal achievement
Variable BSE B b
Step 1
Feedback 0.55 0.27 0.15*
Step 2
Strategy systematicity 0.64 0.22 0.21**
Step 3
Feedback 0.35 0.28 0.09
Strategy systematicity 0.55 0.23 0.18*
*p!0.05; **p!0.01.
597R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
influences motivational state. A good argument that this causal relationship has this direction is
that first learners choose a strategy and we related strategy to a subsequent test of motivational
state administered after six trials. This model received an even better model fit, AGFI Z0.98,
CFI Z1.00, RMSEA Z0.001, c
2
(3) Z2.72, pZ0.44, showing that both variables, strategy
systematicity and motivational state can be called mediators for the effect of feedback on
goal achievement, but strategy systematicity affects motivational state. Learners expecting
feedback chose a more systematic strategy, which led to more motivation during learning,
that finally resulted in a better learning outcome.
3.2. Explorative analyses
As we unexpectedly had so many participants finishing after Round 1 we explored how these
26 fast learners differed from slow learners finishing after Rounds 2 (nZ44) or 3 (nZ141). A
reason to finish early could be that these learners had already acquired enough knowledge so
that they decided to begin with the application phase. We tested this assumption post hoc. If
this assumption is correct participants finishing early should have attained more knowledge
at equivalent stages than had the participants who continued learning. Table 5 shows that
the earlier participants finished, the more knowledge (measured by structure scores) they had
attained at the end of Round 1, F(2, 210) Z119.89, p!0.001. The StudenteNewmaneKeuls
post hoc test showed that all three groups differed from each other. Perhaps this extra
knowledge was gained by these learners using more time to think about the task, however,
the finding that early finishers had more knowledge does not appear to be due to these learners
spending more time learning during Round 1, as there was no difference in time spent on
Round 1 by learners who finished after Round 1 and by those who used two or three rounds,
F(2, 202) Z0.04, pZ0.97).
To check which variable, the motivational state or the strategy systematicity, was the best
predictor for knowledge acquisition we calculated correlations for these three variables for
each individual’s last round for fast (i.e., finishing Round 1) versus slow learners (i.e., finishing
Rounds 2 or 3). For each individual’s last round we analysed the three variables for the round
in which the learner stopped learning. On the basis of Vollmeyer and Rheinberg (2000) we ex-
pected that strategy systematicity would be a better predictor for the structure scores of the fast
learners than it would be for slow learners. For learners who needed more rounds to learn the
task learning depended strongly on their persistence. Therefore we expected motivational state
to be a better predictor for structure score in this group of learners. We found indeed that for the
fast learners (nZ26) strategy systematicity correlated higher with structure score,
r(26) Z0.78, p!0.001, than with motivational state, r(26) Z0.49, pZ0.010. This difference
was statistically significant, zZ2.53, pZ0.016. However, for the slow learners (nZ184) the
Table 4
Correlations (r,p) between feedback, the mediating variables and goal achievement
Feedback
a
Strategy systematicity Motivational state
Strategy systematicity (final round) 0.30**
Motivational state (final round) 0.11 0.32**
Goal achievement 0.15* 0.21** 0.46**
*p!0.05; **p!0.01.
a
Feedback is coded: 0 Zno feedback, 1 Zfeedback.
598 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
motivational state was a better predictor for structure score, r(184) Z0.53, p!0.001, than
strategy systematicity, r(181) Z0.45, p!0.001, but this difference was not significant.
4. Discussion
Overall we can conclude that our hypothesis that feedback improves performance was con-
firmed. In addition, we found that feedback increased strategy systematicity (Hypothesis 1), but
not the motivation during learning that was measured with the motivational state (Hypothesis
2). Both performance measures eknowledge acquisition and application edemonstrated
that the feedback group performed better in that they learnt more about the system’s structure
(Hypothesis 3), and they could better apply their knowledge about the structure (Hypothesis 4).
According to our cognitiveemotivational model strategy systematicity could be supported as
a mediator through regression analysis. An additional path analysis demonstrated that indeed
both variables can be regarded as mediators (Hypothesis 5).
Most studies on feedback interventions analyse the effects on final performance and have no
measures for the process, that is, how the learners reached their learning outcome. With our
cognitiveemotivational model we could differentiate two mediating variables: strategy system-
aticity and motivational state. We had expected that these two variables would be affected by
the experiences a learner had with the task. Thus the feedback should not have had an impact
until the end of Round 1, when the first feedback was given. In fact, the announcement alone
that they would receive feedback led participants to use a more systematic strategy, even before
they received any actual feedback. Using better strategies improved motivation during learning
Motivational state .46
No Feedback/Feedback .32 Goal achievement
.30
Strategy systematicity
Fig. 2. Path analysis showing the effects of feedback on performance through two mediating variables.
Table 5
Descriptive statistics (M, SD) for how the participants finishing Round 1 (nZ26) differed from participants finishing
Round 2 (nZ44) or Round 3 (nZ141)
Time for Round 1 Structure score in Round 1
Participants finishing
Round 1 1092.88 (753.99) 2.82 (0.66)
Round 2 1115.98 (525.43) 2.20 (0.64)
Round 3 1083.08 (697.55) 0.76 (0.81)
599R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
that then helped knowledge acquisition and application. Thus, the mere expectation of feedback
triggered deep processing of the learning material. This was a surprising effect.
How does this surprising effect influence our cognitiveemotivational model? Our aim for
running this study was to show that feedback has a similar function as initial motivation.
This assumption was not supported. Whereas initial motivation has a strong influence on mo-
tivational state and a weak effect on strategy systematicity (Vollmeyer & Rheinberg, 1999,
2000), it is the opposite for feedback. Even in the first round an announcement of feedback
encouraged participants to choose a more systematic strategy, which then affected motivation
during learning. So there seems to be a difference in the sequence of the variables. However, at
that stage of research we do not want to put too much emphasis on the order of variables
because, first, the measure for motivational state needs improvement and, second, the results
for feedback need replication.
The question arises of how such an effect of expectation could come about. This effect can
be better explained in terms of motivational processes rather than cognitive processes. Using
a systematic strategy requires high cognitive engagement and deeper processing than does
an unsystematic, but fast, trial-and-error strategy. This variation of cognitive engagement is
a product of motivation.
Unfortunately our motivational state measure used only three items and did not capture this
motivational component. The differences between the feedback and no feedback groups on
motivational state were in the expected direction, however, they were not statistically signifi-
cant. The motivational state items asked participants to indicate whether it was fun to do the
task and what they thought was their probability of success, yet the announcement of
feedback seems not to have influenced these motivational components. We suspect that other
motivational components play a role. Maybe the expectation of feedback simply leads to a
higher commitment to do the task, because the learners themselves as well as the experimenter
can find out how well they performed. This speculation, however, has to be tested in new
studies in which commitment is measured before and during the learning. As we did not
expect such a commitment effect we did not measure it (a possible item to measure this could
be: ‘‘I am absolutely determined to perform well on this task.’’). Of course, this is only one of
many possible explanations. It has the advantage that it appears plausible and can easily be
tested.
Although the motivational state was not sensitive to our manipulation, it did correlate with
performance. In accordance with our expectations this effect was stronger for learners who
needed more trials to learn the task because they had not learned the system’s structure in
the first round. We have already reported a similar result (Vollmeyer & Rheinberg, 2000).
In future experiments it will be necessary to first find components of motivation during
learning that are sensitive to the feedback manipulation. Then it will be possible to study
the impact of different kinds of feedback. In this study we used task-related feedback
(Rheinberg, 1980), which is only one out of three modes of feedback. For learning in schools,
beside task-related feedback, there exists feedback based on social comparison and feedback
focusing on intra-individual temporal information (Butler, 1987; Butler & Nisan, 1986;
Rheinberg, 1980). Teachers differ with regard to which kind of feedback they prefer, but
it may have an enormous effect on students’ learning motivation (for an overview, see
Heckhausen, Schmalt, & Schneider, 1985). Based on our study we can now experimentally
study the effects of other feedback variations on the learning process and its outcome.
Our results have implications for educational settings. On the basis of the previous literature
on feedback effects on learning researchers could recommend teachers to use feedback because
600 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
of three reasons: (1) feedback is motivating, (2) feedback gives information, and (3) feedback
satisfies state of affairs (see Kulhavy & Wager, 1993). Our results even go further. They give
hints that announcing feedback improves learning strategies. Maybe learners work more care-
fully, once they know that teachers check their learning outcomes.
Acknowledgements
We would like to thank Bruce Burns for comments on this paper. This research was
supported by DFG Grant Vo 514/5 to Regina Vollmeyer and Falko Rheinberg.
References
Ammons, R. B. (1956). Effects of knowledge of performance: a survey and tentative theoretical formulation. Journal of
General Psychology,54, 279e299.
Anderson, J. R. (1993). Rules of the mind. Hillsdale, NJ: Erlbaum.
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psychological Review,64, 359e372.
Atkinson, J. W. (1964). An introduction to motivation. Princeton, NJ: Van Nostrand.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-
Hall.
Bangert-Drowns, R. L., Kulik, C. C., Kulik, J. A., & Morgan, M. T. (1991). The instructional effect of feedback in test-
like events. Review of Educational Research,61, 213e238.
Baron, R. B., & Kenny, D. A. (1986). The moderatoremediator variable distinction in social psychological research:
conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology,51, 1173e1182.
Burns, B. D., & Vollmeyer, R. (2002). Goal specificity and dual-space search theories of problem solving. The Quarterly
Journal of Experimental Psychology,55A, 241e261.
Butler, R. (1987). Task-involving and ego-involving properties of evaluation: effects of different feedback conditions on
motivational perceptions, interest and performance. Journal of Educational Psychology,79, 474e482.
Butler, R., & Nisan, M. (1986). Effects of no feedback, task-related comments and grades on intrinsic motivation and
performance. Journal of Educational Psychology,78, 210e216.
Funke, J. (1991). Solving complex problems: exploration and control of complex systems. In R. J. Sternberg, &
P. A. Frensch (Eds.), Complex problem solving: Principles and mechanisms (pp. 185e222). Hillsdale, NJ: Erlbaum.
Heckhausen, H., Schmalt, H.-D., & Schneider, K. (1985). Achievement motivation in perspective. New York: Academic
Press.
Jussim, L., Soffin, S., Brown, R., & Ley, J., et al. (1992). Understanding reactions to feedback by integrating ideas
from symbolic interactionism and cognitive evaluation theory. Journal of Personality and Social Psychology,62,
402e421.
Kluger, N., & DeNisi, A. (1996). The effects of interventions on performance: a historical review, a meta-analysis, and
a preliminary feedback intervention theory. Psychological Bulletin,19, 254e284.
Kulhavy, R. W., & Stock, W. A. (1989). Feedback in written instruction: the place of response certitude. Educational
Psychology Review,4,91e98.
Kulhavy, R. W., & Wager, W. (1993). Feedback in programmed instruction: historical context and implications for
practice. In J. V. Dempsey, & G. C. Sales (Eds.), Interactive instruction and feedback (pp. 3e20). Englewood Cliffs,
NJ: Educational Technology Publications.
Kulik, J. A., & Kulik, C. C. (1988). Timing of feedback and verbal learning. Review of Educational Research,58,
79e97.
Lewin, K., Dembo, T., Festinger, L., & Sears, P. S. (1944). Level of aspiration. In J. McHunt (Ed.), Personality and the
behavior disorders,Vol. 1 (pp. 333e378). New York: Ronald Press.
Niemivirta, M. (2002). Motivation and performance in context: the influence of goal orientations and instructional
setting on situational appraisals and task performance. Psychologia,45, 250e270.
Nolen, S. B. (1996). Why study? How reasons for learning influence strategy selection. Educational Psychology Review,
8, 335e355.
601R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
Rheinberg, F. (1980). Leistungsbewertung und Lernmotivation [Achievement evaluation and learning motivation].
Go
¨ttingen, Germany: Hogrefe.
Schmidt, K. H., & Kleinbeck, U. (1990). Effects of goals and feedback on performance: mediating mechanisms and
structures of information processing. In P. J. D. Drenth, & J. A. Sergeant, et al. (Eds.), Work and organizational,
social and economic, cross-cultural.European perspectives in psychology,Vol. 3 (pp. 55e66). Oxford, England:
John Wiley & Sons.
Thorndike, E. L. (1913). The original nature of man. In: Educational psychology,Vol. I. New York: Columbia
University, Teachers College.
Vollmeyer, R., Burns, B. D., & Holyoak, K. J. (1996). The impact of goal specificity on strategy use and the acquisition
of problem structure. Cognitive Science,20,75e100.
Vollmeyer, R., & Rheinberg, F. (1998). Motivationale Einflu
¨sse auf Erwerb und Anwendung von Wissen in einem
computersimulierten System [Motivational influences on the acquisition and application of knowledge in a simulated
system]. Zeitschrift fu¨r Pa¨dagogische Psychologie,12,11e23.
Vollmeyer, R., & Rheinberg, F. (1999). Motivation and metacognition when learning a complex system. European Jour-
nal of Psychology of Education,14, 541e554.
Vollmeyer, R., & Rheinberg, F. (2000). Does motivation affect performance via persistence? Learning and Instruction,
10, 293e309.
Weinert, F. E., & Kluwe, R. H. (1987). Metacognition, motivation, and understanding. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Woodworth, R. S., & Schlosberg, H. (1954). Experimental psychology. New York: Holt, Rinehart, and Winston.
602 R. Vollmeyer, F. Rheinberg / Learning and Instruction 15 (2005) 589e602
... Social feedback is known to play an important role in human interactions. Studies showed that participants who received feedback about the execution of a task performed better [12], more so with negative feedback than with positive feedback [13][14][15]. The reason could be that people interpret positive feedback as an indication that their strategy is adequate and negative feedback that they need to update their strategy [15]. ...
... It is worth noting that participants knew that the robot could provide feedback in this task. Vollmeyer and Rheinberg [12] suggested that feedback expectation itself could improve performance. Moreover, in our experiment, two out of three blocks with no social feedback came after blocks with negative feedback. ...
Article
Full-text available
Trust is fundamental in building meaningful social interactions. With the advance of social robotics in collaborative settings, trust in Human–Robot Interaction (HRI) is gaining more and more scientific attention. Indeed, understanding how different factors may affect users’ trust toward robots is of utmost importance. In this study, we focused on two factors related to the robot’s behavior that could modulate trust. In a two-forced choice task where a virtual robot reacted to participants’ performance, we manipulated the human-likeness of the robot’s motion and the valence of the feedback it provided. To measure participant’s subjective level of trust, we used subjective ratings throughout the task as well as a post-task questionnaire, which distinguishes capacity and moral dimensions of trust. We expected the presence of feedback to improve trust toward the robot and human-likeness to strengthen this effect. Interestingly, we observed that humans equally trust the robot in most conditions but distrust it when it shows no social feedback nor human-like behavior. In addition, we only observed a positive correlation between subjective trust ratings and the moral and capacity dimensions of trust when robot was providing feedback during the task. These findings suggest that the presence and human-likeness of feedback behaviors positively modulate trust in HRI and thereby provide important insights for the development of non-verbal communicative behaviors in social robots.
... In their qualitative results, the lack of teacher feedback emerged as a source of demotivation. On the other hand, Vollmeyer and Rheinberg (2005) conclude that, although feedback improves learning strategies and students' final performance, it has no impact on their motivation. ...
... This result is consistent with previous studies conducted by Krishnan and Pathan (2013) and , who found that lack of feedback can lead to L2 demotivation. However, the findings contradict Vollmeyer and Rheinberg's (2005) findings that, although feedback is important for enhancing students' learning strategies and final performance, it has no positive effect on motivation. ...
Article
This study explores demotivational teaching practices, such as teachers’ behaviours, teaching methods, personality traits, and competencies to examine how these practices negatively affect students’ motivation to learn foreign languages. The study utilised a qualitative approach in which semi-structured interviews were conducted with 20 female students in a public secondary school in Almadinah, located in the western region of Saudi Arabia. The study identified five teaching practices that secondary students found demotivating: (a) teacher-centred teaching; (b) excessive use of Arabic; (c) teachers’ lack of motivation; (d) lack of learning feedback; and (e) excessive strictness. Therefore, English teachers should avoid these practices and raise awareness of the detrimental effects they can have on students’ motivation.
... Large repositories of questions that cover all relevant concepts are required in order to support effective practice [26]. Given the critical role that feedback plays in learning [27,28], the presence of model solutions is also important for providing immediate feedback to learners which can prompt reflection and promote selfregulated learning behaviours [29]. Furthermore, research on the problem description effect has shown that the presence of relevant contextual information in the wording of a question can impact cognitive load and have a positive effect on problem success [30,31]. ...
Preprint
Full-text available
In this article, we introduce and evaluate the concept of robosourcing for creating educational content. Robosourcing lies in the intersection of crowdsourcing and large language models, where instead of a crowd of humans, requests to large language models replace some of the work traditionally performed by the crowd. Robosourcing includes a human-in-the-loop to provide priming (input) as well as to evaluate and potentially adjust the generated artefacts; these evaluations could also be used to improve the large language models. We propose a system to outline the robosourcing process. We further study the feasibility of robosourcing in the context of education by conducting an evaluation of robosourced and programming exercises, generated using OpenAI Codex. Our results suggest that robosourcing could significantly reduce human effort in creating diverse educational content while maintaining quality similar to human-created content.
... One common approach for providing feedback in introductory programming courses is the use of automated assessment systems [2,35,60], which at the minimum provide feedback on the correctness of programming assignments submitted for evaluation. As feedback plays a considerable role in learning [32], in addition to influencing approaches to learning by simply being offered [86], it should be given with care; feedback can both improve self-efficacy and decrease self-efficacy [32]. In general, formative feedback -feedback given as a part of the learning process -is preferred over summative feedback, i.e. feedback given after the learning process [40,77]. ...
... This would indicate that the majority of undergraduate and postgraduate dental students in this study were pleased with feedback delivered to them, which was perhaps against the trend of students from other faculties in the UK [7]. In contrast, others have reported that student learners expecting feedback, would use superior approaches from the very beginning, to facilitate their learning [48]. This is encouraging as the majority of student respondents in this study clearly received feedback and responded to it positively, thereby advancing their ongoing learning. ...
Article
Introduction Feedback from teachers to students plays an important role in informing students about the outcome of their assessments. It contributes to students’ ongoing learning. The aim of this study was to investigate dental students’ perceptions of the feedback given to them by their teachers in Europe. Materials & methods An online questionnaire was completed by dental students throughout Europe in this quantitative study. Data were collected via Google Forms, transferred to an excel spreadsheet and analysed using SPSS software Version 24. Results 234 students studying in 9 different European countries completed the questionnaire. These students were born in 36 different countries within and beyond Europe. 84% (n=197) were undergraduate students. 20.3% (n=48) students reported receiving feedback following summative assessments. 81.2% (n=190) students reported constructive criticism as their preferred mode of receiving feedback. 11.3% (n=26) students did not know who delivered the feedback to them. 71% (n=166) students felt that the feedback they received had a significant impact on their future learning. Conclusion It would appear that there is some diversity in dental students’ perceptions of: i) who delivers feedback, ii) when feedback is given, iii) the consistency of feedback received, and iv) the style of feedback they preferred compared to that delivered by tutors. Feedback is being provided to dental students in an appropriate and helpful manner, although there is still room for improvement. Students were aware of the significance of feedback and its impact on future learning.
... One common approach for providing feedback in introductory programming courses is the use of automated assessment systems [2,35,60], which at the minimum provide feedback on the correctness of programming assignments submitted for evaluation. As feedback plays a considerable role in learning [32], in addition to influencing approaches to learning by simply being offered [86], it should be given with care; feedback can both improve self-efficacy and decrease self-efficacy [32]. In general, formative feedback -feedback given as a part of the learning process -is preferred over summative feedback, i.e. feedback given after the learning process [40,77]. ...
Preprint
This article explores the natural language generation capabilities of large language models with application to the production of two types of learning resources common in programming courses. Using OpenAI Codex as the large language model, we create programming exercises (including sample solutions and test cases) and code explanations, assessing these qualitatively and quantitatively. Our results suggest that the majority of the automatically generated content is both novel and sensible, and in some cases ready to use as is. When creating exercises we find that it is remarkably easy to influence both the programming concepts and the contextual themes they contain, simply by supplying keywords as input to the model. Our analysis suggests that there is significant value in massive generative machine learning models as a tool for instructors, although there remains a need for some oversight to ensure the quality of the generated content before it is delivered to students. We further discuss the implications of OpenAI Codex and similar tools for introductory programming education and highlight future research streams that have the potential to improve the quality of the educational experience for both teachers and students alike.
Article
Full-text available
В статье представлено исследование влияния различных форматов обратной связи об академических достижениях на оценку их студентами в условиях дистанционного и обычного обучения. Анализ выявил, что наиболее состоятельными являются теоретические модели, которые учитывают процессы интерпретации учащимися сообщений обратной связи, исходящих от педагога. В эмпирическом исследовании показано следующее: 1) студенты, которые позитивно оценивают обратную связь, более успешны в освоении учебного курса «Психология общения»; 2) в условиях дистанционного обучения преимущество в глазах студентов имеет текстовый формат обратной связи по сравнению с видеообращением преподавателя; 3) в условиях обычного обучения выступление педагога перед студентами имеет большее преимущество в глазах студента над обратной связью в формате текста. Различия оценок обратной связи разных форматов в условиях «онлайн» и «офлайн» обучения обусловлены различным сочетанием стратегий переработки сообщений педагога (систематической и эвристической), которые используют студенты. В целом эти данные свидетельствуют в пользу конструктивистских моделей обратной связи, согласно которым сообщения от педагога подвергаются активному переосмыслению, и его результаты отражаются на реальных учебных достижениях. Значимым фактом, полученным в исследовании, является зависимость оценок различных форматов обратной связи от того или иного режима обучения. В этой связи понимание сущности механизмов воздействия обратной связи возможно только при учете фактора контекста ее введения.
Article
This study is one of the first research attempts to examine students’ satisfaction with web-based assessment platforms. Using the self-determination theory, the research model examines the intermediate role of autonomy and competency. We gathered our data from a survey of 324 business students at one of the Midwestern American institutions who used McGraw-Hill Connect. PLS-SEM analytical procedures were used for testing and validating the hypotheses. The data show that competency, autonomy, quality, and feedback have a significant impact on students’ satisfaction. Functionality, quality, and feedback are the antecedents to autonomy and competence. The relationship between functionality and satisfaction is fully mediated by competency. Implications for research and practice are discussed.
Chapter
This chapter explores the application of feminist theory to online grading feedback processes in higher education. Christianakis describes teacher research as a feminist act. This chapter presents the argument that grading feedback can be viewed from a complementary lens. Many believe that feedback, when offered correctly, has the ability to transform a learner for the better. When actively and intentionally viewed from a feminist perspective, feedback offers a plethora of opportunities to not only teach content, but also to empower and address power inequities. Feminist theory also offers helpful guidance on how to do so.
Article
It is widely believed that feedback improves behavior, but the mechanisms behind this improvement remain unclear. Different theories postulate that feedback has either a direct effect on performance through automatic reinforcement mechanisms or only an indirect effect mediated by a deliberate change in strategy. To adjudicate between these competing accounts, we performed two large experiments on human adults (total N = 518); approximately half the participants received trial-by-trial feedback on a perceptual task, whereas the other half did not receive any feedback. We found that feedback had no effect on either perceptual or metacognitive sensitivity even after 7 days of training. On the other hand, feedback significantly affected participants' response strategies by reducing response bias and improving confidence calibration. These results suggest that the beneficial effects of feedback stem from allowing people to adjust their strategies for performing the task and not from direct reinforcement mechanisms, at least in the domain of perception.
Book
Full-text available
Die Monografie dokumentiert die Entwicklung des Konzepts der Bezugsnorm-Orientierung (BnO). Lehrer unterscheiden sich darin, ob sie die Leistungen ihrer Schüler bevorzugt im interrindividuellen Vergleich, also z.B. im Querschnittvergleich der Leistungen in einer Schulklasse oder ob sie bevorzugt den einzelnen Schüler mit sich selbst vergleichen, also im intraindividuellen Längsschnitt seiner bisherigen Leistungsentwicklung. Aus dem Wechselspiel zwischen theoretischen Annahmen, Beobachtungen der Unterrichtspraxis und empirischen Befunden kristalisierte sich sukzessive das motivierungsbedeutsame Variablensystem aus Leistungsvergleich, Ursachenerklärung von Schülerleistungen, Erwartungsbildung, Sanktionierungsstrategie und Individualisierungstendenz heraus, das das BnO-Konzept bildet. Die Monografie stellt die einzelnen Schritte dieser Forschung dar und berichtet über schülerseitige Motivationseffekte dieser Lehrervariable. Das Buch ist seit vielen Jahren vergriffen. Allerdings bewirkt die hohe Leistungsheterogenität in integrierenden bzw. inkludierenden Schulklassen eine bleibende Aktualität der Bezugsnorm-Thematik bei der schulischen Leistungbeurteilung. Von daher schien es angezeigt, den Text wieder verfügbar zu machen, der die Grundlage bildete für viele nachfolgende Projekte, Weiterentwicklungen und Praxisempfehlungen.
Article
Full-text available
Forty-eight students participated in an experiment in which they had to learn the structure of a complex, computer-simulated system. Subjects manipulated the system and they could analyze the result. By doing so they could increase their knowledge about the system, that is, they could learn the relations underlying the system. In the subsequent application phase they could use this knowledge to reach specified goal values for the system. Performance was measured both in terms of the number of correct relations ("structure score", a measure of knowledge acquisition in the learning phase) and how close subjects got to the goal values ("solution error", a measure of knowledge application). We examined the impact on two motivational factors ("interest in the task" and "confidence of success vs. fear of failure") on both performance measures and by what processes these influences may be mediated. A path-analysis showed a path in which a more systematic strategy leads to more knowledge in the learning phase, which in turn leads to better performance in the application phase. However, the motivational factor "confidence of success vs. fear of failure" had a direct impact and two indirect influences on knowledge application. The two indirect impacts on performance were mediated by strategy systematicity, and "effortless concentration" respectively.
Article
Full-text available
In this article, we attempt to distinguish between the properties of moderator and mediator variables at a number of levels. First, we seek to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating, both conceptually and strategically, the many ways in which moderators and mediators differ. We then go beyond this largely pedagogical function and delineate the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena, including control and stress, attitudes, and personality traits. We also provide a specific compendium of analytic procedures appropriate for making the most effective use of the moderator and mediator distinction, both separately and in terms of a broader causal system that includes both moderators and mediators. (46 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Our cognitive–motivational process model (Vollmeyer, R. & Rheinberg, F. (1998). Motivationale Einflüsse auf Erwerb und Anwendung von Wissen in einem computersimulierten System. [Motivational influences on the acquisition and application of knowledge in a simulated system.] Zeitschrift für Pädagogische Psychologie, 12, 11–23.) assumes that motivational factors affect performance via mediators. Such a potential mediator is persistence. Fifty-one students learnt a complex dynamic system. We measured their initial motivation (mastery confidence, incompetence fear, interest and challenge), then a set of mediating variables during learning. Performance measures were knowledge acquisition and knowledge application. A path-analysis showed that initial motivation influenced persistence. However, any possible relationship between persistence and performance was disrupted because learners with more knowledge stopped sooner (i.e., were classified as lowly persistent). Thus highly persistent learners did not have higher mean final performance, despite acquiring more knowledge over trials.
Article
Feedback is an essential construct for many theories of learning and instruction, and an understanding of the conditions for effective feedback should facilitate both theoretical development and instructional practice. In an early review of feedback effects in written instruction, Kulhavy (1977) proposed that feedback’s chief instructional significance is to correct errors. This error-correcting action was thought to be a function of presentation timing, response certainty, and whether students could merely copy answers from feedback without having to generate their own. The present meta-analysis reviewed 58 effect sizes from 40 reports. Feedback effects were found to vary with control for presearch availability, type of feedback, use of pretests, and type of instruction and could be quite large under optimal conditions. Mediated intentional feedback for retrieval and application of specific knowledge appears to stimulate the correction of erroneous responses in situations where its mindful (Salomon & Globerson, 1987) reception is encouraged.
Article
Feedback is an essential construct for many theories of learning and instruction, and an understanding of the conditions for effective feedback should facilitate both theoretical development and instructional practice. In an early review of feedback effects in written instruction, Kulhavy (1977) proposed that feedback’s chief instructional significance is to correct errors. This error-correcting action was thought to be a function of presentation timing, response certainty, and whether students could merely copy answers from feedback without having to generate their own. The present meta-analysis reviewed 58 effect sizes from 40 reports. Feedback effects were found to vary with control for presearch availability, type of feedback, use of pretests, and type of instruction and could be quite large under optimal conditions. Mediated intentional feedback for retrieval and application of specific knowledge appears to stimulate the correction of erroneous responses in situations where its mindful (Salomon & Globerson, 1987) reception is encouraged.
Article
A meta-analysis of findings on feedback timing and human verbal learning showed that a variety of results have been obtained in 53 separate studies of the topic. Applied studies using actual classroom quizzes and real learning materials have usually found immediate feedback to be more effective than delayed. Experimental studies of acquisition of test content have usually produced the opposite result. Laboratory studies of list learning have produced a variety of results, but the variation in results seems to be related systematically to features of the studies.