Reading motivation as a mediating factor for the effect
of a bibliotherapeutic training on reading
comprehension
By:
Wikke J. van der Putten
Name: Wikke J. van der Putten
Studentnr: 10371796
Words: 4.239
Date: 24-04-2015
Supervisor: Gorka Fraga González
University of Amsterdam
Abstract
This study focuses on the mediating role of reading motivation for the effect of the #BOOK training on reading comprehension. Recent studies have shown that reading
motivation is an important predictor for reading comprehension. Therefore, it is expected that reading motivation will have a mediating role in the training effect on reading comprehension. This is studied by comparing a control group with a training group. Before and after the training, reading comprehension and reading motivation was measured and compared with an ANCOVA. Results showed a significant training effect for reading comprehension and motivation. Additionally, a mediation analysis was performed for reading motivation. No significant mediators were found. It can be concluded that #BOOK is an effective training program, but reading motivation is not a mediating factor in the training effect.
Introduction
Being able to read adequately is essential in our society, since reading comprehension is an important ability for academic success, future employment and personal wellbeing (Ng, Bartlett, Chester, & Kersland, 2013). Reading comprehension is often defined as the ability to understand, use and reflect on written texts (Becker, McElvany, & Kortenbruck, 2010). Learning to read is a complex process that requires several skills. Decoding skills and vocabulary are examples of important predictors for reading comprehension (Becker et al., 2010). Another important predictor of reading comprehension is reading motivation. A study of Stokmans (2006) has shown that the relationship between reading motivation and reading behavior is stronger than between reading skills and reading behavior. In the Netherlands, however, reading attitude is substantially lower than the global and European average (Twist, Gnaldi, Schagen & Morrison, 2004). It has been well established that there is a drop in
reading motivation when children go to secondary school, and that the amount of reading declines substantially in that period of time (Broekhof, 2013). Since reading comprehension is such an important skill, it is important to try to improve reading comprehension and reading motivation. That is why in this study, the effectiveness and mechanism of an intervention, the #BOOK training, that focuses on reading comprehension and reading motivation will be studied.
Several studies have examined the relationship between reading motivation and reading comprehension. There seems to be a bidirectional relationship between reading comprehension and reading motivation (Morgan & Fuchs, 2007). Children who are more motivated, read more often and in turn become more skilled readers. However, poor readers are less motivated to read and therefore read less. The construct ‘motivation’ is divided into two components: intrinsic motivation and extrinsic motivation. Intrinsic reading motivation refers to when a person reads for their own pleasure and when they are interested in reading. In contrast, extrinsic reading motivation is when a child only reads for rewards or for external recognition (Becker et al., 2010). One study looked at the relationship between reading motivation and reading comprehension, accounting for differences between intrinsic and extrinsic motivation (Becker et al., 2010). That study found that intrinsic reading motivation in Grade 4 was positively related to reading ability in Grade 6. However, extrinsic reading motivation was negatively related to reading ability. In addition, another study showed that intrinsic reading motivation predicted the amount of reading more strongly than extrinsic reading motivation (Wigfield & Guthrie, 1997). This suggests that it may be important to focus on improving intrinsic reading motivation instead of extrinsic reading motivation. A study of Schaffner and Ulferts (2013) confirmed that children who read more often, have better reading comprehension. Reading amount also predicts growth in reading
reading can also be seen as an important predictor of reading comprehension. Thus, when trying to improve reading comprehension, it might be best to focus on improving reading motivation. A child with an improved reading motivation will read more and this will improve their reading comprehension.
In the present study, students who live in areas with a low socio-economic status (SES) participated. This is because some studies suggest that children from families with low incomes have lower reading abilities (Greene & Anyon, 2010). Furthermore, it has been shown that children with a low SES also have low reading motivation (Baker, Scher, & Mackler, 1997). Thus, the situation at home can be influence the child’s reading
comprehension and reading motivation. An explanation is that children with a low SES are less stimulated to read by their parents than children with higher SES (Baker et al., 1997). There are less books at home and this makes reading less accessible. Another factor that might be important is the language that is spoken at home. Many parents who originate from for example Turkey or Morocco, probably speak less Dutch at home. Students with a Turkish or Moroccan background score lower in reading comprehension than Dutch students
(Stokmans, 2006). Therefore, this group has a risk of developing problems in reading
comprehension, which will have an effect on their later academic life. That is why it would be important for an intervention to focus on improving reading comprehension and reading motivation, but also to be effective for students with a low SES.
There is already a number of trainings that have appeared to be effective in improving reading comprehension for students with a mixed SES. One example is a training of
metacognition and working memory that improved reading comprehension (Carretti,
Caldarola, Tencati, & Cornoldi, 2014). Another training was aimed at enhancing the strategies of children when reading a text. This also had a positive effect on reading comprehension (Sung, Chang, & Huang, 2008). However, the literature shows that reading motivation is a
very important factor to focus on when trying to improve reading comprehension (Becker, Wigfield en Fuchs). That is why it would be useful to design a training that focuses on improving reading motivation. An example of a training program that focusses on improving reading motivation is bibliotherapeutic training, which that has been successfully
implemented in the United States (Polleck, 2007; Polleck, 2010). This training program was subsequently edited and implemented in the Netherlands and is called #BOOK. In this training, students read books that they can choose themselves. They read the book at home and discuss it at school in small groups guided by a trained psychology or orthopedagogic Master student. The books are about characters with whom the students can identify
themselves and who are confronted with problems and challenges. By discussing the books, students are being stimulated to make up their own opinion about the problems that the characters face. There is no pressure from school, meaning that the students do not have to take a test after reading the book and they do not receive any grades for the training. In this way they read for their own pleasure instead of because they have to do so. This will enhance their intrinsic reading motivation and by improving this, it is assumed that reading
comprehension will also be improved. The training could also have a direct effect on reading comprehension.
In the present study the effectiveness of the bibliotherapeutic training #BOOK and the mechanism of the training will be studied. If the training is effective, it is also important to know in which way the training works. This information is useful for improving the training. The question this paper aims to answer is: Is reading motivation a mediating factor for the effect of the #BOOK training on reading comprehension. The following hypotheses were tested: [1] Training increases reading comprehension. [2] Training enhances reading motivation. [3] Reading motivation predicts reading comprehension. [4] The initial reading motivation predicts reading comprehension after the training. [5] Enhanced reading
motivation is a predictor for the training effect of reading comprehension. It is expected that training will improve both reading comprehension and reading motivation of students. It is also expected that reading motivation will predict reading comprehension. Finally, it is expected that reading motivation on the pretest and on the posttest will mediate the training effect of reading comprehension. This will be studied with a randomized controlled trial, through which it is possible to compare a training group with a control group to find out if the training is effective.
(1)
(3) (4) (5)
(2)
Figure 1. Model of the influences of the #BOOK training.
Methods
Participants
In the study participated one hundred students from three different secondary schools in Amsterdam, The Netherlands. The students were all in the first year of the preparatory secondary vocational education (in Dutch: Voorbereidend Middelbaar Basis Onderwijs). Most of the students (70%-95%) lived in neighborhoods where the majority of the people have a low SES. The students were randomly assigned to a training condition (n=44) and a control
Reading comprehension
pretest
Reading motivation
pretest Reading motivation post test
Reading comprehension post
condition (n=56). The sample size consisted of Some of them did not complete the pretest or posttest and were therefore not included in the analyses. The final sample consisted of 40 students in the training condition and 50 students in the control condition. There were 66 girls and 24 boys that participated in the study with an average age of 13 years and 6 months, SD = 0.8. The caretakers of the students received an informed consent form, through which they could object to their child participating in the study. No parent objected to their child participating. The students received no reward for participating in the study. The study has been approved by the Ethics Committee of the University of Amsterdam in 2014.
Materials
Reading comprehension was measured with the Vlaamse Test Begrijpend Lezen (VTBL; Flemisch test of reading comprehension; van Vreckem, Desoete, de Paepe & van Hove, 2010). In this test the comprehension of a narrative text is measured. The test consists of a text with . First the student has to read the text carefully without a time limit. After reading the text, the student receives two questions that they have to answer without looking at the text, these are the memory questions. After filling in those questions the student is allowed to look at the text when answering the rest of the questions. All the question are multiple choice with four answer alternatives. A correct answer is worth one point. Because the memory questions were not reliable, they were excluded from the analysis. The student could receive a maximum of 24 points in this test. A score of 24 on the test means a really high level of reading comprehension. A low score on the test corresponds with a low reading comprehension. The reliability of the test (Cronbach’s alpha) is r = .78 (van Vreckem et al., 2010).
Reading attitude is measured with the Gobale Leesattitude (Global reading attitude), which was derived from De Casus Bazar (Stokmans, 2007). This questionnaire consists of 21 items, which are all pairs of words. The students have to state which word matches their opinion about reading in his spare time on a five-point scale. The items are divided into hedonistic reading attitude and utilitarian reading attitude. These are comparable with intrinsic reading attitude and extrinsic reading attitude, respectively. Hedonistic reading attitude is pleasure in reading and utilitarian reading attitude is the instrumental value of reading. An example of a hedonistic item is interesting versus not interesting. An example of an utilitarian item is no waste of time versus waste of time. The maximum score for hedonistic reading attitude is 50 and the maximum score for utilitarian reading attitude is 55. Therefore, the total maximum score is 105. A low score matches with high reading attitude. The higher the score, the lower the reading attitude. The internal consistency for both scales is good (Cronbach’s Alpha: r = .94 and r = .83; Stokmans, 2007).
The public library of Amsterdam provided the books for the project. Trainers could select books and when the students decided which book they preferred to read, the library would lend the books to the trainers.
Procedure
The training was implemented from March until June in 2014. The training was given weekly at school and one training session took 45 minutes. Because of holidays and school activities, it was not possible to provide a training session every week, but all the students followed eight to ten sessions. The students could pick books from a selection made by the public library of Amsterdam, resulting in a difference in books that each group of students read. Students that were in the control condition followed their regular Dutch class. Before and after the training there was a pretest and a posttest. The measures took around 45 minutes and were taken during school hours in a classroom.
Master students psychology and orthopedagogy gave the training. The trainers received a course about the training to teach them about the theoretical and practical aspects of the training. They also received a protocol about the execution of the training and there were regular meetings to discuss the training. By doing this, intervention fidelity was monitored.
Statistical analysis
For the evaluation of the effects of training, an analysis of covariance (ANCOVA) has been used, since this is a good way to examine the training effects of a randomized controlled trail (Vickers & Altman, 2001). The conditions were the independent variable, the posttest was the dependent variable and the pretest was taken into account as a covariate. A Pearson correlation was used to analyze if reading motivation on the pretest predicts reading
comprehension on the pretest. The pretest of reading motivation was the independent variable and the pretest of reading comprehension was the dependent variable. It is expected that when reading motivation improves, reading comprehension will also improve. Therefore the
Pearson correlation was reported one sided. For the mediation analysis, Hayes PROCESS plugin (Hayes, 2013) has been used with condition as the independent variable, reading comprehension on the posttest was the dependent variable and reading comprehension on the pretest was the covariate. The following mediators were included: Hedonistic reading attitude, utilitarian reading attitude and total reading attitude, for all the pretest and the posttest were used.
Results
For the analyses, 90 students were used. The distribution between girls and boys was checked for equality with a Chi-squared test. The distribution was equal between the
conditions, X2= .409, p = .522. Age was tested with and independent t test and it was shown that the groups did not differ in age, t(78) = -.736, p= .464. Based on this, it can be concluded that the groups do not differ from each other so they can be compared. No assumptions for the ANCOVA’s were violated.
Table 1
Means and standard deviations of descriptive values for the training condition and control condition.
Training condition Control condition Mean (SD) Mean (SD)
Total number 40 50
Male 12 12
Female 28 38
Age 13;7 (0;8) 13;6 (0;8)
In order to analyze the effect of the training on reading comprehension, an ANCOVA was performed with the posttest scores as dependent variable, the condition as independent variable and the pretest scores as a covariate. An effect for condition was found, F (1,86) = 6.71, p< .050, ƞ2 =0.72. By looking at figure 2, it can be seen that the score on the reading comprehension test improves for students in the training condition but declines for students in the control condition. The means and standard deviations can be found in table 2.
0 2 4 6 8 10 12 14 16 18 20 training control VT BL S co re Condition pre post
Figure 2. Mean scores of the VTBL test for reading comprehension on the pretest and posttest
divided by condition
Table 2
Means and standard deviations of the scores on the Bazar test for reading attitude divided for training condition and control condition.
Condition Pretest mean (SD) Posttest mean (SD)
Reading comprehension
Training 12.73 (4.38) 13.80 (4.90)
Control 13.14 (4.52) 12.10 (5.35)
Hedonistic reading attitude
Training 33.36 (7.97) 29.21 (8.37)
Control 31.51 (8.87) 33.20 (8.35)
Utilitarian reading attitude
Training 31.46 (7.09) 30.69 (6.55)
0 10 20 30 40 50 60 Training Control Sc or e B azar Condition
Hedonistic reading attitude
Pretest Posttest 0 10 20 30 40 50 60 Training Control Sc or e B azar Condition
Utilitarian reading motivation
Pretest Posttest
Total reading attitude
Training 64.82 (14.01) 59.90 (14.15)
Control 63.27 (17.24) 65.84 (16.89)
In order to analyze the effect of the training on reading attitude, an ANCOVA was also performed with the total score on the Bazar test and the different subscales. An effect for condition was found for the total score of reading attitude, F (1, 85) = 5.89, p< .050, ƞ2 = .06. For hedonistic reading attitude also an effect of condition was found, F (1, 85) = 9.33, p< .050, ƞ2 = .10. However, no effect of condition was found for utilitarian reading attitude, F (1, 85) = 1.41, p= .119, ƞ2 = .02. As can be seen in figure 3, hedonistic reading attitude and total reading attitude seem to improve in the training condition but not in the control condition. For the utilitarian reading attitude no clear effect can be seen in the figure. The means and
standard deviations are shown in table 2. In sum, students in the training condition improved more than the control condition for hedonistic reading attitude but there was no difference in utilitarian reading attitude.
0 20 40 60 80 100 Training Control Sc or e B azar Condition
Total reading attitude
Pretest Posttest
Figure 3. Means of the scores on the Bazar test for reading attitude, hedonistic reading
attitude and utilitarian reading attitude on the pretest and the posttest divided by training condition and control condition.
The relation between reading comprehension and reading attitude was analyzed by executing a Pearson correlation between reading comprehension and hedonistic reading attitude, utilitarian reading attitude and the total score of reading attitude. There is a significant negative relationship between reading comprehension and reading attitude, r = -.18, p< .050. This indicates that when reading comprehension improves, reading attitude also improve. There is a negative correlation because a low score on the Bazar test, means a high level of reading attitude. There is also a negative correlation between reading comprehension and hedonistic reading attitude, r =-.17, p< .050 and between reading comprehension and utilitarian reading attitude, r = -.17, p< .050. The correlations between reading comprehension and the different components of reading attitude are low, but there are correlations.
For the mediation analyses, the PROCESS function by Hayes was used (Hayes, 2013). The mediating factors that were included in the models are the pretest and the posttest for hedonistic reading attitude, utilitarian reading attitude and total reading attitude. Indirect
Condition b= -.24 p= .23 Direct effect: b= -2.43, p< .050 indirect effect: b= .18, [-.038-.89] Hedonistic reading attitude pretest Reading comprehension posttest Reading comprehension pretest b= -.089 p= .081 b= -2.07 p= .25 b= .72 p< .001
effects were made up an used to find significant mediators. When a 95% CI interval has a 0 in it, the mediator is not significant. In Figure 4, the mediation models are shown with
corresponding estimates, significant levels and 95% CI intervals. There was no significant indirect effect of reading comprehension before and after the training through hedonistic reading attitude on the pretest, b = 0.18, BCa CI [-0.038, 0.89] and also not through hedonistic reading attitude on the posttest, b = -0.23, BCa CI [-0.87, 0.083]. There was no significant indirect effect of reading comprehension before and after the training through utilitarian reading attitude on the pretest, b = -.014, BCa CI [-0.47, 0.34] and also not through utilitarian reading attitude on the posttest, b = -0.16, BCa CI [-0.78, 0.099]. There was no significant indirect effect of reading comprehension before and after the training through total reading attitude on the pretest, b = -0.25, BCa CI [-0.90, 0.027] and also not through total reading attitude on the posttest, b = 0.10, BCa CI [-0.19, 0.63]. In sum, none of the studied mediators had a significant effect on the training effect of reading comprehension.
Total reading attitude pretest Reading comprehension pretest Reading comprehension posttest b= -.36 p= .34 b= -.045 p= .10 Direct effect: b= -2.00, p< .050 indirect effect: b= -.25, [-.90-.027 Condition b= -1.94 p= .57 b= .72 p< .001 Utilitarian reading attitude pretest Reading comprehension pretest Reading comprehension posttest b= -.13 p= .94 b= -.1022 p= .0450 Direct effect: b= -2.23, p< .050 indirect effect: b= -.014, [-.47-.34] Condition b= .74 p< .001 b= -.20 p= .33 Hedonistic reading attitude posttest b= -.17 b= -.062 p= .23 b= 3.73 p= .037
Utilitarian reading attitude posttest Reading comprehension pretest Reading comprehension posttest b= -.19 p= .35 b= -.091 p= .069 Direct effect: b= -2.09, p< .050 indirect effect: b= -.16, [-.78-.099] Condition b= 1.71 p= .35 b= .74 p< .001 Total reading attitude posttest Reading comprehension pretest Reading comprehension posttest b= -.44 p= .25 b= -.054 p= .046 Direct effect: b= -2.35, p< .050 indirect effect: b= .105, [-.1916-.63] Condition b= 5.44 p= .11 b= .74 p< .001
Figure 4. Mediation models for the training effect of reading comprehension, with estimates,
significance levels, direct effects and indirect effects.
Discussion and conclusions
In this study, it was investigated if reading attitude is a mediating factor for the effect of a bibliotherapeutic training on reading comprehension. After the #BOOK training was implemented, effects of reading comprehension, hedonistic reading attitude and total reading attitude were found. The students who were part of the training improved their reading comprehension and hedonistic reading attitude. There was a low correlation found between reading comprehension and reading attitude, therefore it cannot be concluded that reading attitude is an important predictor of reading comprehension. Finally, reading attitude on the pretest and posttest were both no significant mediators in the training effect of reading comprehension.
A possible explanation for the disappointing results on the mediation analysis, is the sample that was used. In this sample, students lived in areas with a low socio-economic status (SES). The studies that did found strong relationships between reading comprehension and reading attitude, used samples with a mixed SES (Morgan & Fuchs, 2007; Wigfield &
Guthrie, 1997).Students with a low SES have lower reading comprehension and reading motivation (Baker et al., 1997; Greene & Anyon, 2010). It is a possibility that because this group has a lower level of reading comprehension and reading motivation, the relationship between these factors are different than in other groups. To investigate if this is the case, students with a normal SES should receive the training, to compare them with the students with a low SES.
Another interesting component that could be added to the model is reading amount. By enhancing the exposure to books, reading comprehension will become better (Schaffner & Ulferts, 2013). Perhaps the students in the training condition improve their reading
comprehension because they read more. Reading amount could be a mediating factor for the training effect of reading comprehension. In further research, students can be asked to keep track of the hours they read. It would be expected that students will read more when they are in the training condition.
The positive results of the training, however, are promising for a new effective intervention. The training, as expected, had a positive effect on reading comprehension and reading attitude. Even though the first results are promising, there are still ways to improve the training and by doing so, the training effects. This can be done by giving more extensive training to the students that provide the training. In this way, it can be made sure that the focus will be on the right components such as reading motivation and reading comprehension. An extended protocol for the training can help with this. This will also make sure that the training is the same for all students who follow the training.
There are also aspects of the study that can be improved. In this study, only a pretest and a posttest were used. The results of an effective intervention should still be seen after a few months. A follow up measure moment would be necessary to know if the effects of the training remain. Another improvement would be to have more participants in the study, this
will make the conclusions stronger. Even though the #BOOK training seems like a good intervention, it could also be a method of prevention. When children go to secondary school, their reading motivation drops (Broekhof, 2013). By implementing an intervention before their motivation drops, this drop in reading motivation can be prevented. In this study, the intervention is implemented when they have just started secondary school. It could be possible, that when children have the training when they are still in primary school, the drop in reading motivation might be prevented.
In conclusion, this study showed that the #BOOK training is a promising new
intervention that improves reading comprehension and reading motivation for students with a low SES. There is still more research needed to find out how the training works exactly. When there is more information about the #BOOK training, it is possible to improve it and implement it at schools. By doing this, students with a low SES can improve their reading abilities and with that, perhaps also their academic future.
Literature
Baker, L., Scher, D., & Mackler, K. (1997). Home and family influences on motivations for reading. Educational Psychologist, 32, 69–82.
Becker, M., McElvany, N., & Kortenbruck, M. (2010). Intrinsic and extrinsic reading
motivation as predictors of reading literacy: A longitudinal study. Journal of Educational
Psychology, 102(4), 773–785.
Broekhof, K. (2013). Meer lezen, beter in taal – vmbo: Effecten van lezen op
Carretti, B., Caldarola, N., Tencati, C., & Cornoldi, C. (2014). Improving reading
comprehension in reading and listening settings: The effect of two training programmes focusing on metacognition and working memory. British Journal of Educational
Psychology, 84(2), 194–210.
Cipielewski, J., & Stanovich, K. E. (1992). Predicting growth in reading ability from
children’s exposure to print. Journal of Experimental Child Psychology, 54(1), 74–89.
Greene, K., & Anyon, J. (2010). Urban School Reform, Family Support, and Student Achievement. Reading & Writing Quarterly, 26(February 2015), 223–236.
Morgan, P. L., & Fuchs, D. (2007). Is There a Bidirectional Relationship Between Children’s Reading Skills and Reading Motivation? Exceptional Children, 73(2), 165–183.
Ng, C.-H. C., Bartlett, B., Chester, I., & Kersland, S. (2013). Improving Reading Performance for Economically Disadvantaged Students: Combining Strategy Instruction and
Motivational Support. Reading Psychology, 34, 257–300.
Schaffner, E., & Ulferts, H. (2013). Reading amount as a mediator of the effects of intrinsic and extrinsic reading motivation on reading comprehension. Reading Research
Quarterly, 48, 369–385.
Sung, Y. T., Chang, K. E., & Huang, J. S. (2008). Improving children’s reading comprehension and use of strategies through computer-based strategy training.
Computers in Human Behavior, 24(4), 1552–1571.
Wigfield, A., & Guthrie, J. T. (1997). Relations of children’s motivation for reading to the amount and breadth or their reading. Journal of Educational Psychology, 89(3), 420– 432.
