The Effect of Bibliotherapy on Reading Skills and
Social-Emotional Skills among Students following Intermediate
Vocational Education
Manon Bannink 10001847
First assessor: Gorka Fraga Gonzalez Second assessor: Jurgen Tijms External supervisor: Mirthe Stoop
Index Abstract 3. Introduction 3. Reading skills 4. Vocabulary 5. Reading comprehension 6. Reading attitude 7.
Need for proper reading education 8.
Social-emotional skills 11.
Bibliotherapy 13.
Summary 16.
Research design 17.
Method
Design & Analyse 18.
Participants 19. Materials 20. Procedure 22. Results Analysis posttest 22. Single-case-analysis 24. Qualitative Questionaire 29.
Conclusion & Discussion 30.
Abstract
This study examined the effect of bibliotherapy on reading skills and social-emotional skills. The factors vocabulary, reading comprehension, reading attitude and social-emotional skills were specifically measured. This research included first grade students following intermediate vocational education, who were divided in a training group, which followed bibliotherapy, and a control group, that did not follow bibliotherapy. Due to a high dropout of participants only 7 participants out of the 87 participants from both groups were left to
analyse if we used the original design. In which the results of the participants would have been analysed by an ANCOVA using the posttest as dependant variable, the group condition as independent variable and the pretest as the covariable. As an alternative approach, only the results of the posttest were analysed assuming that the groups did not differ from each other on in performance on the pretest. This analysis showed a positive effect of bibliotherapy on vocabulary and reading attitude. Also a few participants from the training group were analysed individually. These results showed a positive effect of bibliotherapy on reading comprehension and reading attitude. No support was found for the expected positive effect of bibliotherapy on social-emotional skills.
Introduction
From the age of 18, students no longer fall within the public education law. This makes dropping out of school easier. School dropout can be described as leaving school before graduating. In the Netherlands the number of school dropouts was around 25,9701 in 2013/2014. The European Parliament sees dropping out of school as a severe problem and tries to reduce the number of school dropouts. The Dutch government has also taken precautions to reduce the amount of school dropouts. Several precautions have been taken from a severe absenteeism policy to personal assistance in studying for students
(Rijksoverheid, 2015). The last few years the number of school dropouts has been reduced from 39,600 to 25,970, this is respectively a reduction from 3.2% in 2013/2014 to 1.9% in 2013/2014. The government’s goal is to reduce the number of school dropouts to less than 25,000 in the years 2015 and 2016 (Rijksoverheid, 2015). Especially in intermediate
vocational education the amount of school dropouts is very high. Despite reducing the number of school dropouts, the percentage of school dropouts among students following intermediate vocational education still was 5.2% average in 2013/2014. Some schools giving intermediate vocational education even had a percentage of 7% school dropouts. The government has set
the goal to reduce the amount of school dropouts in intermediate vocational education to 5% on average in 2016.
Zooming in on the possible causes for school dropout, several studies have shown that school success has an influence on school dropout. Accordingly, less to none success in school leads to a higher risk of school dropout (Broekhof, 2013; Hernandez, 2011). According to the findings of these studies, focussing on enhancing school success will lead to a reduction in school dropouts. Research shows that school success can be predicted by the level of reading skills and the level of social-emotional skills (Blum & Libbey, 2004; Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011; Farrell et al., 2006; Hernandez, 2011; Webster-Stratton & Reid, 2004). Most students dropping out of school come from
disadvantaged communities. A study from Van Bijsterveld-Vliegenthart (2012) shows that the amount of school dropouts in disadvantaged communities is twice as high compared to more advantaged communities. In line with the theory, studies show that students from
disadvantaged communities lack reading skills and social-emotional skills (Baker et al., 1997; PISA, 2009; Broekhof, 2014). Based on these findings, it seems likely that improving reading skills and social-emotional skills can lead to an enhancement in school success. Higher school success can, in its turn, lead to a reduced risk of school dropout. Previous studies in the United States have shown that bibliotherapy has a positive effect on reading skills and social-emotional skills (Polleck, 2007; 2011; Sridhar & Vaughn, 2002; Sullivan & Strang, 2002). In the Netherlands a project based on the bibliotherapy of Polleck and colleagues, has been implemented among students following lower vocational education. The project is called #BOOK. A previous study that was focused on the effects of #BOOK found the same positive effects of bibliotherapy as the study by Polleck and colleagues did (Tijms, Stoop, & Polleck,
in preparation). The present research is based on the same project, #BOOK, but focuses on a different target group: students following intermediate vocational education.
Reading Skills
Reading skills can be divided in different competencies. Previous studies on the effects of bibliotherapy investigated the following competencies: vocabulary, reading comprehension, reading motivation and reading attitude (Tijms, Stoop, & Polleck, in preparation). The results of that research showed a positive effect of bibliotherapy on three out of the four investigated competencies: vocabulary, reading comprehension and reading attitude. Based on these findings, the present research will focus on the reading competencies vocabulary, reading comprehension and reading attitude.
Vocabulary
In brief, vocabulary can be described as a construct consisting of two dimensions:
breadth and depth (Anderson & Freebody, 1981). Breadth refers to the size of one’s vocabulary or the amount of words someone knows. Depth refers to how well someone knows a word: superficial or extensive. It reflects to the quality of one’s vocabulary. More depth consists not only out of knowing the meaning of a word, but also possible other meanings (homonym), other words with the same meaning (synonyms), the orthography, the grammar, the pronunciation etc. Together breadth and depth form someone’s vocabulary (see figure 1).
Figure 1
Breadth and depth dimension of vocabulary, with some words as example
Research shows that new words, when they occur in different contexts, lead to a more depth vocabulary than when they occur without a context or in the same context (Tannenbaum, Torgesen & Wagner, 2006). Reading is a way to get in touch with words in different contexts. In line with this theory Cain, Oakhill and Lemmon (2004) found that a higher reading
frequency leads to a better vocabulary. Cain and colleagues ascertained that the context was used to understand the meaning of a word (better). Extending the vocabulary by reading also provides a more depth vocabulary. As described above reading is a comprehensive way of extending the vocabulary that leads to a more breadth and more depth vocabulary. In contrast to other exercises, such as only linking a word to a definition, which are more focused on only extending the breadth of vocabulary.
Mol and Bus (2011) studied the influence of reading in spare time on reading development and language development in children. Their study showed that the influence of reading in spare time on vocabulary increases with age. That means that the importance of reading in spare time in order to extend the vocabulary gets increasingly important over the years. Besides having a positive effect on vocabulary, reading in spare time also has a positive effect on reading comprehension (Mol & Bus, 2011). In addition to the study from Mol and Bus (2011) several other studies showed that an extensive vocabulary is positively correlated with reading comprehension (Alderson, 2000; Read, 2000; Qian & Schedl, 2004). Qian (2002) adds to the findings that a more breadth and more depth vocabulary relates to a better reading comprehension.
Reading Comprehension
Reading comprehension is described by Tijms and Van Gelder (2008) as the
interpretation of the text by using the theme of the text, the text structure, the key concept of the text, other concepts within the text and the relationship between these concepts and the key concept. The text structure seems to be important to distinguish the main theme from the sub-themes within the text. The text structure refers to the division of the text into paragraphs. The text structure gives the text a better arrangement. To comprehend the text it is important to be able to distinguish relevant information from less relevant information. On top of that, it is also important to be able to make a connection between this important information and the key concept of the text (Tijms & Van Gelder, 2008). In addition Duke and Pearson (2002) note that in order to comprehend the text one has to be able to integrate new information from the text to current knowledge, opinions and experiences. To achieve a good interpretation of the text, not only skills related to reading comprehension are of great importance, also general cognitive skills and basic reading- and language skills are necessary. The level of mastering these skills is related to the level of one’s ability to make an accurate interpretation of the text (Pasquarella Gottardo & Grant, 2012 ). As with vocabulary, it turns out that reading
frequency affects reading comprehension. Reading more leads to better reading comprehension (Mol & Bus, 2011).
From the moment students move on from primary education to secondary education, they are considered to be able to independently integrate information from texts.
A study by Alfassi (2004) shows that texts are considered to be primary sources for the acquisition of knowledge. This shows how crucial reading comprehension is to acquire academic knowledge (Cromley & Azevedo , 2007; Daggett & Hasselbring, 2007; Reschly,
2010). A weak reader is therefore in a disadvantage compared to a stronger reader when it comes to acquiring academic knowledge. A study by Schaffner, Schiefele and Ulferts (2013) shows that reading comprehension can be predicted by reading attitude.
Reading Attitude
Attitude is described by Fishbein and Ajzen (1975) as “A learned predisposition to respond in a consistently favourable or unfavourable manner with respect to a given object”. In the case of reading attitude the ‘object’ in the definition above regards reading. Reading attitude can be divided in two components: hedonistic reading attitude, which regards the pleasure in reading, and utilitarian reading attitude, which regards the utility in reading (Stokmans, 2006). A positive reading attitude among students makes them more motivated to read (Van Schooten, Beater, & Chair, 2004). It is important to note that the motivation in this context is intrinsic, because only intrinsic, not extrinsic, motivation is considered to have a positive effect on reading. Intrinsic motivation can be explained as drive based on internal factors, for example because someone likes doing something. Extrinsic motivation involves drive based on external factors, like getting a reward in terms of a high grade on a test. It has been found that an increased intrinsic motivation is positively correlated to reading
comprehension, in contrast to extrinsic motivation, which is negatively correlated to reading comprehension (Logan, Medford, & Hughes, 2011). A positive reading attitude can be created by having a positive reading experience (Polleck , 2010). Multiple positive experiences over a longer period of time will lead to cultivation of reading behaviour in the positive direction. Eventually, this positive cultivation of reading results in improvement of reading
comprehension (McKenna, Conradi, Lawrence, Jang, & Meyer, 2012). So it appears that reading frequency is also important when it comes to changing reading attitude in the positive direction, similar as for extending vocabulary and improving reading comprehension. This relation is bidirectional. So a positive reading attitude leads to a higher reading frequency and a higher frequency in reading leads to a more positive reading attitude as long as the reading experience is positive.
A main part of one’s reading attitude is formed by the home environment. When parents have a positive view on reading and encourage their children to read, it is more likely that they will develop a positive reading attitude. Children with a high socioeconomic
background often develop a positive reading attitude. In contrast, children with a low socioeconomic background are more likely to develop a negative reading attitude (Van Schooten, De Klopper & Stoel, 2004). As highlighted before, reading plays an essential part
in the acquisition of academic knowledge. Reading behaviour is therefore very important. Having a positive reading attitude will automatically lead to more intrinsic motivation to read academic literature compared to having a negative reading attitude. This is supported by a research from Stockmans (2006). She found that the hedonistic component (pleasure in reading, thus intrinsic) was the strongest predictor of reading behaviour, when comparing reading attitude in general, utilitarian reading attitude and hedonistic attitude to each other. In line with the theory that reading behaviour influences acquiring academic knowledge,
Schaffner, Schiefele en Ulfers (2013) also found that there is a relation between reading attitude and the level of reading comprehension. They found that a negative reading attitude results in the development of reading comprehension in a negative direction.
The Need for Proper Reading Education
An international research from Twist and colleagues (2004) shows that Dutch adolescents belong to the least motivated readers of the world. Zooming in on the reading behaviour of students, it seems that a big shift in reading behaviour occurs when the transition from primary education to secondary education takes place. When students enter secondary education, student’s reading attitude decreases (Guthrie & Davis, 2003; McKenna et al., 2012; Unrau & Schlackman, 2006) and the amount of students that reads with resistance or does not read at all increases (Conradi, Jang, Bryant, Craft & McKenna, 2013; Lenters, 2006).
Notably, most of the students developing a negative reading attitude have a low
socioeconomic background (Baker, Scher, & Mackler, 1997; PISA, 2009). Claassen and Mulder (2006) found that the majority of the students with a low socioeconomic background are following lower vocational education (70% in native adolescents and 80% in foreign adolescents). On a yearly basis around 35.000 adolescents with a low socioeconomic background enter lower vocational education. Research from Broekhof (2014) shows that, when comparing students who follow primary education to those who follow lower vocational education, reading frequency and pleasure in reading are both lower among students
following lower vocational education (see figure 2.1 and 2.2). In 2014, the inspection of education also concluded that most students following lower vocational education have a negative reading attitude. Reading attitude is an important predictor of reading
comprehension, therefore students following lower vocational education might also have trouble with reading comprehension. This is supported by several studies showing that students with a low socioeconomic background often have a negative reading attitude and
also lack of reading comprehension (Park, 2011; PISA, 2009; Van Diepen, 2007; Van de Grift, 2008).
Figure 2.1
Pleasure in reading in primary education and lower vocational education
Note: PE = primary education, LVE = lower vocational education (Source: Broekman, 2014)
Figure 2.2
Reading frequency in primary education and lower vocational education
Note: PE = primary education, LVE = lower vocational education, pxy = times per year, pxm = times per month, pxw = times per week and day = daily (Source: Broekman, 2014)
In order to enhance school success among students that form a high risk of dropping out of school, it seems to be useful to change their reading behaviour. Previous studies, as described above, point out the importance of reading frequency on reading skills, regarding vocabulary, reading comprehension and reading attitude. Increasing reading frequency turns out to have a positive effect on all of these types of skills. Vocabulary and reading attitude both affect reading comprehension. Reading comprehension, on its turn, has an effect on school success.
Similarly to reading comprehension, reading attitude also influences school success (see figure 3).
Figure 3
Interaction reading frequency, readings skills and school success
Especially students with a low socioeconomic background form a group at high risk of dropping out of school. This increased risk is linked to the high rates of a negative reading attitude and poor reading skills among this group (Park, 2011; PISA, 2009; Van Diepen, 2007; Van de Grift, 2008). This leads to a decrease in school success, which in its turn, increases the risk of school dropout (Guthrie, Klauda & Morrison, 2012; Prater, 2006). Compared to strong readers, weak readers are disadvantaged when it comes to interpreting texts, making assignments and following lessons, (Broekhof, 2013). Having a negative reading attitude enlarges this effect even more. In addition, Guthrie & McRae (2012) found that the reading attitude of students with a low socioeconomic background has a more predictive value on school success, than the reading attitude of students with a high socioeconomic background has. This means that school success of students with a low
socioeconomic background depends more on reading attitude, than school success of students with a high socioeconomic background. These findings indicate the importance of changing reading behaviour in a positive direction among students at risk, which are often students with a low socioeconomic background following lower or intermediate vocational education.
Not only home environment influences one’s reading behaviour. School can be a part of developing a positive reading behaviour as well. For example, it appears that it is more
likely for students to develop a positive reading attitude when there is more room to read in class (Van Schooten, De Klopper & Stoel, 2004; Krashen, 1993). It seems that social inequality in reading skills can be reduced by spending more time on reading education, which gives students a more equal chance in school success (Guthrie & McRae, 2012).
Social-Emotional Skills
Besides reading skills, social-emotional skills also have an influence on school success. Students that have less social-emotional skills are also less engaged in school. The decrease in school engagement has a negative effect on school success, which in its turn increases the risk of dropping out of school (Battin-Pearson et al, 2000; Blum & Libby, 2004; Durlak et al., 2011, Farrell et al., 2006). Possessing good social-emotional skills makes one able to communicate and work well with others, behave responsible and respectfully, and take responsible decisions considering problems and challenges (Association for Supervision and Curriculum Development, 2007). Social-emotional skills can be developed by
social-emotional learning (SEL). SEL can be defined as “the process of acquiring core competencies to recognize and manage emotions, set and achieve positive goals, establish and maintain positive relationships, make responsible decisions and handle interpersonal situations
constructively” (Elias et al, 1997; Shechtman & Yaman, 2012, p. 547). Willis and colleagues (2006) add that SEL is important for the personal development, considering developing personal well-being, positive relations with others, effective communication skills, sensitivity towards other’s perspectives and feelings, empathy, respect for other’s point of view and making responsible and well considered decisions. Bandura (1997) states that
social-emotional skills are developed by observation, imitation and learning from social role models, like parents, siblings, peers and teachers.
Multiple researchers showed that students with well developed social-emotional skills are better in managing daily challenges, problems or other stressful situations, compared to students with less social-emotional skills (Ayotte et al., 2003; OECD, 2010; European
Commission, 2006). Well-developed social-emotional skills turn out to have a positive effect on school performance, because they have a positive influence on cooperation and
communication and give an insight into one’s performances (strengths and weaknesses). They also influence school performance, because well-developed social-emotional skills provide more emotional control, which leads to a higher concentration and less distraction in class (Elias, 2006). Social-emotional skills do not only influence school performances and school engagement, they also play a major part in being able to adapt to society. Well-developed
social-emotional skills make one able to work and communicate in a good way with people from different backgrounds and with different cultures. Besides, they make it possible to make democratic and well considered decisions involving other’s perspectives and opinions (Benson, Leffert, Scales, & Blyth, 2012). Apart from the importance of academic skills, social skills also turn out to be important in making a career. Think about doing a job interview for example or being able to work and communicate well with colleagues (Ayotte et al., 2003; OECD, 2010; European Commission, 2006).
Social-emotional learning is possible in many different ways. Programs used for SEL focus on five different competencies: self-awareness, social awareness, responsible decision-making, self-management and relationship management. Self-awareness is the ability to identify and recognise one’s own emotions, knowledge, norms, values, strengths and weaknesses. Social awareness refers to recognising and respecting other’s perspectives, opinions and feelings. Responsible decision making includes the ability to identify and evaluate problems or situations causing problems, being able to evaluate different solutions and reflect on one’s own actions. Self-management can be described as being able to achieve personal goals by controlling and monitoring one’s emotions, impulses and behaviour. The last competence, relationship management, contains the ability to work and negotiate well with others and support them in order to promote a positive relationship (Casel, 2015). Social-emotional skills turn out, as described above, to be important when it comes to forming and maintaining good social relationships and improves school performance. However, a study by Durlak and colleagues (2011) shows that a lot of students lack social-emotional skills. Only 23 to 40 percent of students between 12 and 14 years old said they mastered social-emotional skills like: coping strategies, interpersonal competencies, intercultural competencies, taking perspective, setting goals, making decisions etc. Social emotional skills are not being tested in school or measured on a nationwide basis. It could be that there is less attention focused on social-emotional skills, because the lack of time and/or control (in forms of testing) in schools. In contrast to cognitive skills, such as math and spelling, which are being tested (nationwide). There is recognition for the lack of education in social-emotional skills. The Onderwijsraad (2014) states that there is not enough time devoted to the education of 21st century skills, which social-emotional skills are part of.
By means of a good SEL-program students are taught through systematic
instructions, practice and modelling in which way social-emotional skills can be applied. It becomes habit to apply social-emotional skills, when students get more often in touch with
the competencies on which social-emotional skills are based (Ladd & Mize, 1983; Weissberg, Caplan & Silvo, 1989; Durlak et al. 2011). A SEL-program does not only improve social-emotional skills, but it also improves school performance and reduces behavioural problems (Shechtman & Yaman, 2012). Multiple studies in the United States confirm the presumed positive effects of SEL-programs. The meta-analysis of Durlak and colleagues (2011) shows that students who attended a SEL-program were more engaged in school, had more improved school performances and their general well being improved more compared to students who had not attended a SEL-program. The study also shows that SEL-programs decrease risk-taking behaviour. Which means SEL-programs can be used as precaution according to external and internal behaviour, like aggression, substance abuse or depression.
When students with different backgrounds are compared, it appears that students with a low socioeconomic background form a high risk of dropping out of school. Ayotte and colleagues (2003) confirmed that most of the students with a low socioeconomic background lack social-emotional skills. So, also seen in this perspective, students with a low
socioeconomic background form a group at risk. Previous section noted that students with a low socioeconomic background are vulnerable to failing in school, because of lacking reading skills, in particular: lacking vocabulary, lacking reading comprehension and having a negative reading attitude. The present section explains the low success in school among students with a low socioeconomic background based on lacking social-emotional skills. As explained before, lacking social-emotional skills decreases school engagement. Less engagement leads to a decrease in school success, which in its turn increases the risk of dropping out of school. So it seems that school success is influenced by both reading skills and social-emotional skills.
Bibliotherapy
Making changes in reading behaviour and reading skills by means of reading education and improving social-emotional skills by means of a SEL-program, decreases the social inequality in school success. These measurements are seen as effective precautions against school dropout (EU, 2012; European Commission, 2007; Guthrie & McRae, 2012; Inspectie van het Onderwijs, 2007, 2014). The need for SEL-programs and proper reading education is high (Onderwijsraad, 2011). Especially in schools with al lot of students with a low socioeconomic background who form a high risk of dropping out of school, like lower and intermediate vocational education (Onderwijsraad, 2011).
Social-emotional learning is possible in many different ways. One way is by reading books. SEL by means of reading books works according to Dodger-Feder, Lincoln, Coulson
and Hooker (2013) through two different mechanisms. On the one hand it gives the reader a way to simulate the social and emotional experiences from the character in the book, what gives the reader the possibility to value other’s mental representations more. On the other hand, it gives readers the opportunity to build up knowledge about social-emotional rules and learn more about cause-consequence relations through the context of the story.
Social-emotional learning through books can be seen as using stories and characters in books as social role models. McCullis and Chamberlain conclude that SEL through books leads to more self expression, gives an insight in one’s own thoughts, and those of others, and provides several coping strategies. An American SEL-program that applies SEL by reading books is bibliotherapy (Polleck, 2007; 2011). The Dutch version of this program is known as a bibliotherapeutic book club training, called project #BOOK. In this setting, bibliotherapy can be described as reading and discussing stories that relate to the perception of the students who join the training (Elbaum & Vaughn, 2001). Considering the positive effect of well-developed reading skills and social-emotional skills on school success, bibliotherapy seems to be an effective manner to increase school success. This is because bibliotherapy focuses on both reading and social-emotional skills.
Bibliotherapy is given to a small group of 5 to 7 students. Books that are being read connect to the students’ mental representations and backgrounds and level with their reading- and language skills. The readings are weekly discussed in the groups supervised by a trainer. Bibliotherapy deals with the all the five SEL competencies from Casel (2005). This is done by identifying oneself with the character or the story (self awareness), construct meanings and interpretations of the text (social awareness), getting insight in problems that touches one and discussing those problems (self management and relationship management), gaining new norms and values (self awareness and social awareness), gain conscience that others can experience the same problems as you (social awareness) and gain the ability to think of different solutions for those problems (responsible decision making) (McCullis &
Chamberlain, 2013). Apart from teaching social-emotional skills, bibliotherapy tries to let students experience reading in a meaningful and personal way. This is done by creating a non-educational setting and focussing on pleasure in reading. This is important in order to change students’ reading attitude in a positive way (Kush et al., 2005; Stokmans, 2007). Eventually bibliotherapy has three goals: improve self-reflection and social-emotional learning by reading books, stimulate a positive reading attitude and improve vocabulary and reading comprehension (Tijms, Stoop, & Polleck, in preparation).
As described in the theory before, part of the improvement of reading comprehension will be due to the change in reading attitude (Schaffer et al, 2013; Martin-Chang & Gould, 2008; Taboada, Tonks, Wigfield, & Guthrie, 2008). Another point of improvement of reading comprehension lays in understanding the content of the text and questions/instruction about the text. By discussing the books, students learn to think more about the content of the text while they read. They also practice learning to filter key information of the text, what makes is easier to recognize relations between parts of the text and recognizing the theme of the text (Tijms & van Gelder, 2008). The personal connection between the student and the book, can lead to more comprehension and critical-analytical reactions during the meetings (Wilhelm, 2008). The discussions about the book can also lead to further development of skills like reasoning and critical evaluation, which influences reading comprehension in a positive way (Soter, Wilkinson, Murphy, Rudge, Reninger, & Edwards, 2008). By active practice with the text: discussing it and linking the text to one’s own experiences, students work on improving their reading comprehension on two levels. On a superficial level (what happened, where or when etc.) and on a more profound level (what is the consequence of… or why did it happen etc.). Finally, is stated that active practicing with the text and discussing the text, leads to an extension of vocabulary.
One of the goals of bibliotherapy is to let students get in touch with reading in a positive manner, emphasizing reading in a safe and recreational way and discussing the books without giving it an educational vibe. This is accomplished by choosing books that students can relate to and by not forcing any rules on them (they make their own rules together). This makes it possible for students to get a positive reading experience, which is very important in order to change reading attitude in a positive direction (Kush et al., 2005; Stokmans, 2007). An important part of bibliotherapy is the book discussion in the meetings, this can lead to an insight that reading does not have to be something to do solo, but can also lead to social interaction (Nelck-da Silva Rosa & Schlund Bodien, 2004). Getting this insight is important for developing a positive reading attitude, especially if you want to change reading attitude in adolescents (Merga, 2014; McKenna, 2001). As explained before, reading attitude can be divided in two components: hedonic reading attitude and utilitarian reading attitude
(Stokmans, 2006). Bibliotherapy mainly tries to chance the hedonic component (pleasure in reading), because this component is most important for enhancing the intrinsic motivation to read, which has a positive influence on reading comprehension (Schaffer et al, 2013; Martin-Chang & Gould, 2008; Taboada, Tonks, Wigfield, & Guthrie, 2008). Rozalski, Stewart and
Millo (2010) also found that bibliotherapy can have a positive influence on behaviour towards others, for example to fellow students. The role of the trainer is important in order to achieve this change in behaviour among students following the bibliotherapy. The trainer has to be open and involved in the group process and make it possible for each student to speak freely. By creating a safe and trustworthy environment, the relationship between the trainer and the students will grow stronger and will increase school engagement (Durlak et al., 2011; Polleck, 2010).
Summary
Previous sections made clear that students with a low socioeconomic background form a high risk of dropping out of school. This can be explained by the lack of reading skills and lack of social-emotional skills within this group. Decreased school performance and school engagement both lead to a decrease in school success. Decrease in school success, in its turn, increases the risk of dropping out of school (see figure 4).
Figure 4
Influence reading skills and social-emotional skills on school success and school drop out
It also has become clear that the main part of students with a low socioeconomic background follows lower vocational education. 85% of these students continue on to intermediate vocational education (Sondermeijer, 2015). This makes students following intermediate vocational education a vulnerable target group when it comes school dropout, considering the
relation between lacking reading skills and social-emotional skills and the risk of dropping out of school. This is supported by the finding that school dropout is highest among students following intermediate vocational education (Rijksoverheid, 2015). The government’s goal is set to reduce the percentage of school dropout to 5% among students following intermediate vocational education by the end of 2016. Multiple authorities emphasize the importance of mastering reading skills and social-emotional skills in order to succeed in the contemporary society. Methods to improve both of these skills can be found in proper reading education and proper SEL-programs. These measurements are seen as effective precautions against school dropout (EU, 2012; European Commission, 2007; Guthrie & McRae, 2012; Inspectie van het Onderwijs, 2007, 2014).
Bibliotherapy is seen as an effective manner to improve both reading skills and social-emotional skills. Previous research in the United States has shown positive effects of bibliotherapy on reading behaviour and social-emotional skills (Polleck, 2011; Sridhar & Vaughn, 2002; Sullivan & Strang, 2002). A recent study in the Netherlands used a randomized controlled trial (RTC) to measure the effect of bibliotherapy among students following lower vocational education. This study showed a positive effect of bibliotherapy on vocabulary, reading comprehension, reading attitude and social-emotional skills (Tijms, Stoop, & Polleck, in preparation). Research among students following intermediate vocational education has not yet been done. This target group has the same background as the target group of the previous study. The only difference is that our target group already followed lower vocational education and moved on to intermediate education, so they are relatively older (13 to 14 years old in lower vocational education compared to 17 to 18 years old in intermediate vocational education). What kind of effect bibliotherapy on these students has is unclear. That is why this study will focus on the effect of bibliotherapy on reading skills and social-emotional skills among students following intermediate vocational education.
Research Questions
During the second semester of 2014/2015 a research to the effects of the Dutch bibliotherapeutic book club training, #BOOK, took place at ROC of Amsterdam. The goal was to gain more insight in the effects of bibliotherapy among a vulnerable, but relatively older target group compared to target groups that have been researched in the past. The main question of this research was: What is the effect of bibliotherapy on reading skills and social-emotional skills among students following intermediate vocational education? The following hypotheses were tested:
Hypothesis 1: Bibliotherapy extends vocabulary.
Hypothesis 2: Bibliotherapy improves reading comprehension.
Hypothesis 3: Bibliotherapy changes reading attitude in a positive direction. Hypothesis 4: Bibliotherapy improves social-emotional skills.
The hypotheses are based on the theoretical model shown in figure 4. Hypothesis 1 and 2 are based on the moments within the bibliotherapy where students are reading books and are actively aware of the content of the text. Combined with the discussion about the content of the book. This forms the expectation that vocabulary will be extended and that reading comprehension will be improved. Hypothesis 3 is based on gaining a positive experience with reading during bibliotherapy. Expecting the reading attitude to change in a positive direction, especially the hedonic reading attitude. Hypothesis 4 is based on the depth given to the content of the books by connecting it to one’s own experiences and discussing those experiences with the other group members during the weekly sessions.
Method
Design and analysis
The effects of bibliotherapy are measured by a pretest-intervention-posttest design. The software Power & Precision 4 (Borenstein, Rothstein, & Cohen, 2001) was used to compute the sample size, which was based on a medium effect size, one independent variable with two conditions (training group, control group), α = .05 and an R2 of the covariate ranging from 0.35 to 0.55. For a power of at least 0.80 was determined that n = 31 to n = 42 per group was needed. Eventually, five classes of approximately 20 students were chosen for this research. By means of a passive informed consent students could choose not to participate in the project. This study used an open randomized controlled trial design. Half of the students were randomly assigned to the group that would follow the #BOOK training. This group determined the training group. The other half of the students formed the control group and did not follow the #BOOK training, but followed the usual class. The sample is considered to be representative for the target group, because students from different educational backgrounds participated in this research. After the division the pretest took place, followed by starting the training, which was given 10 times during the weekly hour of Dutch. The control group followed the regular Dutch lessons. After a period of ten sessions the posttest took place. The groups were compared to each other on the pretest and the posttest. On the pretest we
training group would score significantly higher than the control group on the different outcome measures.
For analysing the data there has been chosen to use an analysis of covariance (ANCOVA). The condition (two levels: training group and control group) was used as the independent variable, for the depended variable the posttest score was used as outcome variable and the matching outcome of the pretest was used as covariate. This analysis was chosen, because ANCOVAs are considered to be a powerful approach to analyse the effect of trainings in RCTs (Vickers & Altman, 2001). Because there was a high number of dropouts in participants running an ANCOVA on the results of the participants that were left over would have been irrelevant, because the sample was too small. Therefore alternative manners of analysing the data had to be used. The first solution to this problem was to do a nonparametric Mann-Witney analysis on the outcome of the results of the posttest, under the assumption that the groups had not differed from each other on the pretest. The second manner that we came up with was to do some individual analyses of participants that were in the training group and participated on both the pretest and the posttest. Thus to do a single-case analysis. The
program statistic reliability change (Jacobson, & Truax, 1991) was used for the single-case analysis. This program uses the inter-item reliability and the standard deviation to calculate what the minimal differential score between the pretest and the posttest has to be to state that the difference is significant. Furthermore the results of a qualitative questionnaire were also included in in order to find some support for the outcomes of the analyses that were described above.
Participants
Students from five different classes participated in this study. In total, 87 participants were present on the pre- and posttest. The sample contained one class with a Trading & Entrepreneurship educational background, two classes that participated had a Fashion educational background and other two classes were from the educational direction Legal Services. Participants who did not participate on both the pre- and posttest were excluded. Besides the normal dropout of participants, there were four entire classes with participants that had to be left out when the data would have been analysed using the original design. Because the power was cut while students were taking the test, the two Fashion classes were not able to fully finish the pretest. Furthermore, the two Legal classes were not able to take the posttest, because of a miscommunication with the ROC. Due to these reasons a very large number of participants had to be excluded from of this research considering the
original design. Only 7 participants were left to run the analysis on. For these reasons we used an alternative design that allow us to test 45 participants.
Materials
Bibliotherapy
The bibliotherapy as used in this research is known as the bibliotherapeutic book club training #BOOK, but will be referred to as training later on. The training consisted out of ten meetings, in which a small group of participants (4 to 8) read and discussed the content of books. The trainers were volunteers of foundation Senia (an elderly foundation that does voluntary work) and had a course on how to lead the sessions in a proper way according to the guidelines. The books were chosen by a specialist in youth literature of the public library of Amsterdam. The specialist took in consideration that the theme of the book had to connect to the background of the participants. Each group decided their own norms and rules. The group also decided what had to be read for the next meeting. During these meetings there was time to discuss the content of the book, as well as their own experiences and opinions related to the book.
Vocabulary
Vocabulary was measured using the subtest Woordenlijst of the Differentiële
Aanlegtest voor Onderwijs versie A (DAT) (De Wit & Compaan, 2005). The test consisted of
60 words, for each word five possible answers were given and the participant needed to choose the right synonym. For instance: Expert: A) Master B) Inventor C) Advisor D) Champion, E) Specialist. The test was scored by the Commissie Testaangelegenheden
Nederland (COTAN)2 as followed: test construction: good; quality of the test materials: good; quality of the manual: good; norms: sufficient; reliability: sufficient; construct validity:
sufficient; criterion validity: insufficient(Egberink, Janssen, & Vermeulen, 2007).
Reading comprehension
Reading comprehension was measured by a test developed by Van Lankeren in 2006, the Van Lankeren Begrijpendleestest (VLBLT), consisting of a version A and version B. Both tests were constructed in the same way and consisted out of 10 open questions. For instance: ‘Come up with a fitting title for the entire text’. The inter-item reliability of version A was α = .54 and version B had a reliability of α = .62. Both were assessed as sufficient (Van den Brink & Mellenbergh, 1998; Eggen & Sanders, 1993). Version A was used for the pretest and version B was used for the posttest. This decision was made because there was a
short period of time between the pre- and posttest and we had the luxury to choose between two versions. We did not want the participants to give the same answers as on the pretest without even thinking about the text or reading the text, which is essential if you want to measure reading comprehension. The test-retest correlation between the two versions was assessed as sufficient, r = .60, (Van Lankeren, 2006).
Reading attitude
The reading attitude was measured by the self-report questionnaire Globale
Leesattitude, coming from De Casus Bazar (Stokmans, 2007). Each time, the participant was
given a pair of words, for instance: necessary-unnecessary. Between the two options there was given a five-point scale, on which the participant had to score their thoughts about reading in spare-time. In total the participants were given 21 pairs of words. Apart from a global score of reading attitude, the questionnaire also made a difference between hedonistic reading attitude (pleasure in reading) and utilitarian reading attitude (utility of reading). The total score indicated how much of a negative reading attitude someone had, which meant the higher the score; the more negative the reading attitude. The same goes for hedonistic reading attitude and utilitarian reading attitude. The inter-item reliability was assumed reliable, α = .96 en α = .93 (Stokmans, 2007).
Social-emotional skills
The social-emotional skills were measured using the Dutch and shortened version of the Social Emotional Assets and Resilience Scales (SEARS), questionnaire A (self-report for adolescents) (Merrwell, 2011). This test consisted out of 35 items, with different type of statements that related to social-emotional competencies, for instance: ‘I think it is easy to talk to many different people’. Each statement was scored on a 5-point likert scale, where 1 = never and 5 = always. The English version had an internal consistency which was assumed to be consistent, α = .96 (Cohn, Merrell, Felver-Grant, Tom, & Endrulat, 2009).
Qualitative questionnaire
Since this research was commissioned by ROC of Amsterdam, it was important to do a qualitative questionnaire as well in order to found out more about experiences from participants in the training group. The qualitative questionnaire consisted out of 14 items like: ‘I felt comfortable during the meetings’. Participants scored the items on a 5-point likert scale, where 1 = I totally agree and 5 = I totally disagree. By doing this, the circumstances that were essential for the training to succeed could be checked. Besides, this type of qualitative
example, to form possible explanations about why hypothesized effects would or would not show.
Procedure
The pre- and the posttest were both taken during a lesson Dutch at the ROC of Amsterdam. Each class took the tests on a different time within the same couple of weeks. There were 12 weeks between the pretest and the posttest. In each class, the questionnaires were given in a different order. One out of the five classes took the tests digital, the other four classes did the tests on paper. The posttest ended with the qualitative questionnaire, which was only given to the participants of the training group. Participants sat separate from each other and they were not allowed to talk to each other, to make sure the tests were done individually.
Results
As described before, the data would have been analysed by using an ANCOVA, based on the results of the training group and control group on the pre- and posttest, using the pretest as the co-variable. Because of the high number of participants that was lost, only 7 out of the 87 participants completely filled out the pretest and posttest. Running an ANCOVA on the results of the participants that were left over would have been irrelevant, because the sample was way too small. Therefore an alternative manner of analysing the data was used. The first paragraph of this section will show the results of the nonparametric Mann-Witney analysis on the outcome of the posttest on the factors vocabulary, reading comprehension, reading attitude and social-emotional skills, under the assumption that the groups did not differ from each other on the pretest. The second part of this section will show an analysis on individual level of participants that were in the training group and participated on both the pretest and the posttest. So we will do a single-case analysis in the second paragraph. Finally, we will end the result section with the results of the qualitative questionnaire.
Analysis posttest
In this paragraph was tested if the training group had made more progression
compared to the control group regarding vocabulary, reading comprehension, reading attitude and social-emotional skills. In this research participants were randomly selected for each of the groups. Therefore was assumed that the groups did not differ from each other at forehand. This assumption could not be supported by a statistical test. This meant that the results of the analysis on the posttest had to be interpreted carefully. A total of 45 participants attended on the posttest, 11 of them were in the training group and 34 of them were in the control group.
The exact number analysed results differed per test, because unreliable results or missing results were left out of the analysis. Out of the 45 participants 8 of them were male and 37 of them were female. The mean age was 18 and varied from 16 to 24 (see table 1). Within the groups the male/female ratio did not differ (p = .51). This was tested with Fisher’s Exact test (because the groups did not fulfil the assumptions of the chi square test).
Table 1
Descriptive variables
Training group Control group Number of participants N N
11 34
Sex Male Female Male Female
1 10 7 27
Age jr;mnd M SD M SD
17;7 1,54 17;7 1,4
The results of the training group and the control group on the posttest were tested using the nonparametric Mann-Witney U test. The choice to do a nonparametric test was based on the small sample size and the unequal deviation of participants between the two groups (Field, 2013). Table 2 underneath describes the results of the analysis on the posttest per factor.
Table 2
Overview of the statistic values per factor and per group
Condition* N Mean St. Dev. Median U p z-score Effect size r Reading comprehension TG CG 9 25 8.56 6.76 4.67 3.7 8 7 133 .221 0.81 0.14 (small) Social-emotional skills TG CG 5 21 98.2 101.51 14.55 17.29 98 100 51 .475 -0.1 -0.02 (small) Vocabulary TG CG 5 17 32.5 25.03 9.99 8.28 34 26 63,5 .05 1.65 0.35 (medium) Reading attitude TG CG 9 28 38.22 57.93 9 17.11 38 60 42 .001 -2.98 -0.49 (medium) Hedonic reading attitude TG CG 9 28 17.89 28.25 5.16 9.59 17 29,5 44,5 .001 -2.89 -0.48 (medium) Utilitarian reading attitude TG CG 9 28 20.33 29.68 5.05 8.65 21 30 48,5 .003 -2.75 -0.45 (medium)
*TC=training group; CC=control group
The results showed significant difference between the training group and the control group on two different factors. The training group scored higher on vocabulary than the control group,
U= 63.5, z = 1.65, p = .05. The effect size r (Rosenthal, 1991; Field, 2013) was calculated too and a medium effect size on vocabulary was found, r = 0.35. The other factor on which the training group significantly differed from the control group, was reading attitude. The training group scored significantly lower on global reading attitude than the control group, U = 42, z = -2.98, p = < .05, r = 0.49. Where a lower score indicates a more positive reading behaviour. The training group scored also significantly lower on the hedonic component (U = 44.5, z = -2.89, p = < .05, r = 0.48) and the utilitarian component (U = 48.5, z = -2.75, p = < .05, r = 0.45) of reading attitude than the control group. The global reading attitude had a medium effect size and the effect sizes of hedonic reading attitude and utilitarian reading attitude were also medium. The group differences in reading comprehension (U = 51, z = -0.1, p = .475, r = 0.14) and social-emotional skills (U = 113, z = 0.81, p = .221, r = 0.02) did not differ
significantly from each other.
Single-case-analyses
In order to conclude more about the effect of the training among the intermediate vocational education target group, a few participants of the training group were individually analysed. The amount of participants that attended the pre- and posttest was too small to analyse on group level, therefore the choice was made to do an analysis on single case level. For each test, we calculated what differential score was necessary to conclude that the difference between scores of the pretest and posttest was significant. The program statistic
reliability change (Jacobson, & Truax, 1991) was used for this calculation. This program uses the inter-item reliability and the standard deviation to calculate what the minimal differential score has to be in order to state that difference is significant (see table 3).
Table 3
Shows what minimal differential score is needed for a significant difference per test
Test r SD* N Minimal differential score**
VLBLT 0.792 4.996 61 6.32 SEARS-A 0.893 10.187 42 8.28 DAT 0.785 6.440 46 9.24 BAZAR glob 0.96 19.50 443 10.81 BAZAR hed 0.96 10.30 443 5.71 BAZAR uti 0.93 9.24 443 6.78
*DAT 1 outlier; SEARS 2 outliers deleted for Mean/SD **Based on 95%-confidence interval
Note: VLBLT: Van Lankeren Begrijpendleestest; Sears-A: Social Emotional Assets and Resilience Scales; DAT: Differentiële Aanlegtest voor Onderwijs; BAZAR glob: global reading attitude; BAZAR hed: hedonic reading attitude; BAZAR uti: utilitarian reading attitude
The scores of five different participants of the training group were analysed in this paragraph. The test results of these participants were used to analyse whether there was a significant difference between the results on the pretest and posttest or not. In total only one participant that attended the training group completely filled out all of the tests, as is showed in table 4 underneath. Table 4 does not only show the number of tests that the participants completely fulfilled, it also gives an overview of the differential score between the pretest and the posttest per test per participant. Under each test is showed how much the differential score at least has to before a significant difference can be stated. When comparing the differential scores of the participants to the minimal differential scores needed for a significant difference, 5 out of the 18 scores were significantly different. Two participants scored significantly higher on reading comprehension on the posttest compared to the pretest. One participant scored significantly lower on all the components of reading attitude, when comparing the pretest to the posttest. Table 4
Overview of the differential scores per test and per participant Name (number of tests) Reading comprehension Vocabulary Soc-emo skills Reading attitude glob. Reading attitude hedo. Reading attitude uti. P. H. (4) -2 +1 -3 -3 +1 -4 C. X (2) +3 - - -15 -7 -8 N. D. (2) +16 - - +1 +3 -2 E. S. (2) +7 - - -1 +2 -3 Minimal differential score*: 6.32 8.28 9.24 10.81 5.71 6.78
*Based on a 95%-confidence interval
Participant P.H. was the only participant that completely filled out al of the tests. On none of the factors a significant difference was found within a 95% confidence interval. Only on the factor utilitarian reading attitude a direction toward a significance difference was found. Figure 5.1 shows that the score on the posttest fell within the 68% confidence interval of significance but not yet within a 95% confidence interval. Were we can state that there is a trend toward an effect, but no effect yet.
Figure 5.1
Changes in utilitarian reading attitude within participant P.H.
Within participant C.X. reading comprehension in C.X. seems to get better. Figure 5.2 shows that the score on the posttest of reading comprehension almost falls within the 68%
confidence interval. A difference in reading comprehension could not be found, but a direction towards a trend was noticeable.
Figure 5.2
Figure 5.3, 5.4 and 5.5 show a difference in reading attitude within participant C.X. The scores of global, hedonic and utilitarian reading attitude on the posttest are significantly lower than the scores on the pretest (within a 95% confidence interval).
Figure 5.3
Changes in global reading attitude within participant C.X.
Figure 5.4
Figure 5.5
Changes in utilitarian reading attitude within participant C.X.
Participants N.D. and E.S. both scored significantly higher on reading comprehension on the posttest compared to the pretest. Figure 5.6 and 5.7 show the results of N.D. and E.S. on the pretest and the posttest. There were no significant differences found on reading attitude within N.D. and E.S.
Figure 5.6
Figure 5.7
Changes in reading comprehension within participant E.S.
Qualitative questionnaire
Apart from the quantitative tests, a qualitative questionnaire was taken to check the
manipulation. The questionnaire also measured experiences from the participants in order to give qualitative feedback about the training to the ROC of Amsterdam. A total of 14
participants that followed the training filled out the qualitative questionnaire. Table 5 gives an overview of the subjects of the questions3 and the answers of the participants per item. Table 5
Percent of the participants that agreed, felt neutral or disagreed on an item
Subject Disagree Neutral Agree Missing
Talk 0% 7% 93% -Comfortable 0% 7% 93% -Valuable 0% 21% 79% -Nice 0% 21% 79% -Reading pleasure 0% 36% 64% -Reading comprehension 7% 50% 43% -Taking perspective 7% 64% 29% -Ambiance 14% 57% 29% -Relate 7% 29% 21% 43% Vocabulary 7% 72% 21% -Reacting 36% 43% 21% -Decision making 14% 65% 14% 7%
Free time reading 36% 50% 14%
-Learned about self 50% 36% 14%
-Finding help 36% 64% 0%
Conclusion & Discussion
This research studied the effect of bibliotherapy on reading skills and social-emotional skills among students following intermediate vocational education. Most of the students following intermediate vocational education have a low socioeconomic background.
Literature states that students with a low socioeconomic background often lack reading skills and social-emotional skills compared to students with a high socioeconomic background, which leads to a decrease in school success. Which in its turn leads to a higher risk of dropping out of school and it gives students with a low socioeconomic background a predisposition when it comes to getting a job or making a career. Bibliotherapy seems to be an effective manner to improve reading skills and social-emotional skills. Research among students following lower vocational education shows that bibliotherapy has a positive effect on vocabulary, reading comprehension and reading attitude in particular. Most of the students following lower vocational education continue following intermediate vocational education after they have finished lower vocational education. This means that students following intermediate vocational education often have the same low socioeconomic background as students following lower vocational education. Although, students following intermediate vocational education are relatively older than students following lower vocational education. This raises the question if bibliotherapy will show the same effects on reading skills and social-emotional skills among students following intermediate vocational education as among students following lower vocational education.
Prior to this research was hypothesized that bibliotherapy would have a positive effect on vocabulary, reading comprehension, reading attitude and social-emotional skills. The results in the section above showed that the training group scored significantly higher on the posttest than the control group on the factor vocabulary. There was also a significant differnce on the factor reading attitude. The training group scored significantly lower than the control group on global reading attitude, as well as utilitarian reading attitude and hedonic reading attitude. Where a lower score indicates a more positive reading attitude. This means that after following the training, participants in the training group had a lager vocabulary, experienced more pleasure in reading and also found reading more useful than the ones in the control group. These findings suggest that bibliotherapy had a positive effect on reading attitude and extended the vocabulary among students that had followed the training. It was also
social-emotional skills. However, improvement on these factors was not found based on the results of the group analysis above. Although we used random selection for subscribing participants to one of the two different conditions, it cannot be ensured that the measured differences between the groups in vocabulary and reading attitude can be attributed to following the training, because we could not include the pretest in the analysis. So we could not statistically prove that there was no difference between the groups at forehand. It might have been that the training group already scored higher on vocabulary and reading attitude compared to the control group on the pretest. If that would be the case, the difference between the two groups that was found on the posttest, cannot be attributed to following the training. This is why we decided to run some more tests. We ran a single case analysis on the results from a couple of participants from the training group. The single case analysis was based on the results of the posttest compared to those of the pretest. Within three out of the four
analysed participants positive changes were found. Two participants showed improvement on reading comprehension and one participant showed improvement on all of the components of reading attitude. These results suggest that the training had positive effect on reading
comprehension within two participants and a positive effect of the training on reading attitude within one participant. There were no changes found within any of the participants on
vocabulary or social-emotional skills. Although, it is noticeable that there were no test results of vocabulary and social-emotional skills from three out of the four analysed participants.
When comparing the group analysis to the individual analysis some similarities and some differences were found. In line with the findings on the group analysis, a positive effect of bibliotherapy on reading attitude was found on an individual level as well. The positive effect of bibliotherapy on vocabulary that was founded in the group analysis did not show in the individual analysis. However, a positive effect of bibliotherapy on reading comprehension was found on an individual level in contrast to the group analysis that did not found this effect. The hypothesized positive effect of bibliotherapy on social-emotional skills was not found on any level, not on group level or on individual level. The qualitative questionnaire gives some support to the founded positive effect of bibliotherapy on reading attitude, vocabulary and reading comprehension. For example 64% of the participants said they experienced more pleasure in reading after following the training. The other 36% felt neutral about this matter, which means none of them disagreed. Also 43% said their comprehension in reading got better because of the training and only 7% disagreed on that. As to the
following the training, 21% said they agreed on knowing more words after following the training and only 7% disagreed.
As described above, doing conclusions based on the data of this research is difficult, because a lot of participants had to be excluded. Interpretations based on the results of this research were made carefully and have to be used with the same carefulness in the future. Further research is necessary in order to give more certain conclusions. Taken together three out of the four hypothesized positive effects of bibliotherapy were found. A positive effect of bibliotherapy on reading attitude was found on two levels, on group level and on individual level. This suggests that bibliotherapy changes one’s reading attitude in a positive direction. A positive effect of bibliotherapy on reading comprehension was found on an individual level and a positive effect of bibliotherapy on vocabulary was found on group level. Because the positive effect of bibliotherapy on reading comprehension and vocabulary was only found on one of the two levels it is difficult to draw strong conclusions from.
It is notable that this research only found a positive effect of bibliotherapy on reading skills, but not on social-emotional skills. While previous research, among students following lower vocational education did show a positive effect of bibliotherapy on social-emotional skills. A meta-analysis of Durlak and colleagues (2011) gives a lot of insight in possible factors that could have played a role in the absence of finding a positive effect of bibliotherapy on social-emotional skills. It might be the age of the target group for example. The mean age of the intermediate vocational education student in the current research was 17 to 18 years old, while the mean age of the lower vocational education student, used in
previous reseach to the effect of bibliotherapy on social-emotional skills, was 13 to 14 years old (Tijms, Stoop, & Polleck, in preparation). Most previous research to SEL-programs that found a positive effect of SEL on social-emotional skills had relatively young target groups between the age of 6 and 15 years old (Bierman et al., 2010; Shechtman & Yaman, 2012; Zins, Bloodworth, Weissberg, & Walberg, 2004). The meta-analysis of Durlak and colleagues (2011) shows a positive effect of SEL among a broad target group from 5 to 18 years. This research does not exclude a positive effect of SEL on social-emotional skills in the late adolescence (16-22 years), which would not support the findings of the present study. But their study did found a significantly negative relation between students mean age and skill outcome, which suggests that younger students are more sensitive to the positive effect of SEL on social-emotional skills compared to older students. A difference between the study of Durlak and the present study is the type of SEL-program that is included in the study. Durlak
and colleagues included all type of SEL-programs, while this study only focuses on SEL by bibliotherapy. Also Durlak and colleagues analysed longitudinal data, which the present study did not do. Future research could be focused specifically on the effect of bibliotherapy on social-emotional skills during the late adolescence. Perhaps an effect will show after a longer period of sessions, what would indicate that the positive effect of bibliotherapy on social-emotional skills might become more persistent with age.
Another potential confound for the absence of a positive effect of bibliotherapy on social-emotional skills could be that the trainers had an effect on the training outcome. Trainers used in this study were at least 65 years old, while trainers used in the previous research to the effect of bibliotherapy were around 20 years old (Tijms, Stoop, & Polleck, in preparation). Previous research shows that the attitude towards older adults is more negative compared to the attitude towards younger adults (Kite, Stockdale, Whitley, & Johnson, 2005; Nelson, 2004). The meta-analysis of Kite and colleagues showed that younger adults were rated less stereotypically, seen as more attractive and more component, and were evaluated more favourable than older adults. It might have been that the great age difference between the trainer and the group obstructed the bonding between the trainer and the students, which can lead to outcomes like not opening up in discussions for example. So it might be possible that there is an effect of the trainer on the outcome of the therapy. This effect is also known as therapeutic alliance. A meta-analysis of Martin, Garske en Davis (2000) shows that the
therapeutic alliance has a medium effect on the therapeutic outcome. From this perspective the absence of a positive effect of bibliotherapy on social-emotional skills could be due to the type of trainer. But why would the effect of type of trainer only influence the absence of an increase in social-emotional skills? Perhaps because the content of the sessions form an important part of the social learning process. The discussions within the sessions are
important for developing awareness, social awareness, responsible decision-making, self-management and relationship-self-management (Casel, 2005). The other studied factors
(vocabulary, reading comprehension and reading attitude) are less directly dependent on the content of the sessions. Increase in reading frequency alone can cause a positive effect on reading skills (see figure 3) and therefore the sessions are not per definition necessary. For the improvement of social skills it is important that discussions about the book flow over into discussions on a more personal level (McCullis & Chamberlain, 2013) during the sessions. When students are less likely to bond with the trainer or aren’t engaged to talk about their personal life, the effect of social-emotional learning within the book club is less likely to
