Smits, Boon, Sluijsmans, and Van Gog (2008) conclude that in a web-based learning
environment, high-achieving students benefit more from global EF than from elaborate EF.
Remarkably, students reported that they appreciated the elaborate feedback more. In this
experiment, global feedback contained KCR and solution steps. Elaborate EF included KCR,
worked-out solution steps and accompanying explanations. No effects were found for the lowachieving
students or regarding feedback timing.
Previous research showed that there is no guarantee students will pay attention to
feedback (e.g., Stobart, 2008; Timmers & Veldkamp, 2011). In the selected studies,
researchers seemed to assume that students pay attention to the feedback that is presented.
Only one study examined the time students spent on reviewing material for both correct and
incorrect responses (Morrison, Ross, Gopalakrishnan, & Casey, 1995). They found that the
amount of time students spend on reading feedback is influenced by incentives. As well, the
group of students who spent the most time reading the feedback also performed highest. In the
other studies, the amount of attention paid to feedback by students was not taken into
consideration. However, in six studies, the time students spent completing the entire
assessment was logged (Clariana & Lee, 2001; Corbalan, Kester, & van Merriënboer, 2009;
Corbalan et al., 2010; Kopp, 2008; Lee et al., 2010; Park & Gittelman, 1992).
The majority of the studies were conducted at universities (n = 10). The rest were
conducted at high schools (n = 5), in vocational education settings (n = 2), or at a college (n =
1). None of the studies was conducted in primary education settings. No relationship was
found between the type of education and feedback effects.
The purpose of this study was to gain insight into effective methods for providing
written feedback in a CBA and to identify gaps in current knowledge of this topic. In
analysing the results, a distinction was made between feedback types (Shute, 2008), levels
(Hattie & Timperley, 2007), timing (Shute, 2008), and the level of learning outcomes (Bloom
et al., 1956; Gagné, 1985).
Of the 1158 studies found in the literature search, only 18 satisfied the inclusion
criteria and some additional quality criteria. These outcomes clearly suggest that there is not
much high-quality research available yet on this topic. Furthermore, of the 18 studies
analysed, only nine found one feedback condition to be significantly superior to another. As
well, only seven studies provided usable results for this study: In the studies performed by
Wang et al. (2006) and Wang (2007), no effects could be assigned to a specific type of
feedback. These studies do suggest, however, that providing students with certain feedback
interventions can be beneficial for learning outcomes.
A limitation of this study—and of systematic reviews in general—is the impossibility
of including all the studies relevant to the research question. This is because by using specific
search strategies, one makes selections that do not necessarily include all relevant studies.
Also, in this study, strict requirements were established regarding the quality of acceptable
This study does not contain enough data to provide evidence on the effects of different
ways of providing written feedback. It does, however, give an indication of what might be