










































ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

13 
 

 

 

 

Meta-Analysis of the Effectiveness of Philosophy for Children 
Programs on Students’ Cognitive Outcomes 

 
 

Sijin Yan, Lynne Masel Walters, Zhuoying Wang, and Chia-Chiang Wang 
 
 

ABSTRACT: Philosophy for Children (P4C) is an educational program that aims at introducing 
philosophy into K-12 education. This meta-analysis examines the research on P4C, published from 2002 to 
2016, regarding how it affects pre-collegiate students’ cognitive outcomes. Ten studies (including two 
follow-up studies) with the total sample size of 1,509 students from second to twelfth grade are included in 
this meta-analysis. Results suggest that the extant empirical studies on P4C show an overall moderate 
positive effect (d=0.58) on students’ cognitive learning outcomes and a significant positive effect on 
reasoning skill (d=1.06). Specifically, those studies conducted in non-Western countries have higher effect 
sizes than the Western ones. Moreover, studies with smaller sample sizes have higher effects sizes than 
those with larger sample sizes. This may be because P4C produces better outcomes in reasoning skills than 
general cognitive abilities and reading comprehension, and P4C could be more effective when practiced in 
small scales.   

 
Keywords: Meta-analysis; Philosophy for Children; Cognitive Outcomes; Reasoning Skills 
 

hilosophy for Children (often abbreviated as P4C) is an educational program that provides 
students in K-12 settings opportunities to engage in communities of philosophical inquiry with 
the long-term aim of improving their cognitive abilities (Lam, 2012; Trickey & Topping, 2004). 

An increasing number of studies have documented the implementation of P4C and its impacts on 
students’ cognitive outcomes (Abbasi & Ajam, 2016; Lam, 2012; Nia, 2014). This meta-analysis aims 
to examine what the cumulative research evidence suggests about how P4C affects students’ cognitive 
abilities and whether characteristics of interventions, students, or outcome types influence the 
magnitudes of the effectiveness of the program.  

Introduction of Philosophy for Children 

Philosophy for Children is an educational program initiated by Matthew Lipman, Ann Sharp 
and their colleagues in the Institute for the Advancement of Philosophy for Children (IAPC) in the 
early 1970s (Brandt, 1988; Lam, 2012; Marashi, 2008; van der Straten Waillet, Roskam, & Possoz, 
2015). Witnessing the weaknesses of college students’ argumentative performance in public discourses 
and the tumultuous political environment during the contentious years of Vietnam War in the 1970s 
(Vansieleghem & Kennedy, 2012), Lipman argued that philosophy should no longer be confined to 
college and academic research. Children, he said, even in elementary grades, can begin a quest in 
philosophy to learn how to think and reason (Brandt, 1988).  

P 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

14 
 

 

 

 

 
Currently, with the help of numerous philosophers, educators and researchers, Philosophy for 

Children has become a global movement that has spread across 50 countries, and its material has 
been translated into 20 languages (Daniel & Auriac, 2011). It is now a fertile ground for educators to 
creatively practice different educational ideas (Meskin & Cook, 2012; Murris, 1992) to critically 
modify it for different students living in specific cultural, social and educational contexts in the world 
(Di Masi & Santi, 2016; Ghazinejad & Ruitenberg, 2014; Ndofirepi & Cross, 2015; Ndofirepi & 
Shanyanana, 2016), and most recently, to introduce it to very young children who are below the age 
of 6 (Giménez-Dasí, Quintanilla, & Daniel, 2013; Säre, Luik, & Tulviste, 2016).  

 
As a whole, Philosophy for Children believes in engaging pre-collegiate students in doing 

philosophy by 1) removing the formidable terminologies and 2) using children’s literature, pictures, 
films or other forms of stimuli to bring the philosophical discussion into class.  

 
The central pedagogy of Philosophy for Children is community of inquiry. Its philosophical 

roots can be traced back to the American philosopher Charles Sanders Peirce. In his article Some 
Consequences of Four Incapacities (1868), he claimed that it is pernicious to make single individuals 
absolute judges of truth and what we need is a community of inquiry to ‘grind off the arbitrary and 
the individualistic character of thought’ (Peirce & Houser, 1998). Dewey fleshed out the Peircean 
theory of inquiry and incorporated it into his philosophy of education (Lipman, 2004), contending 
that a genuine community life in the classroom could let students cultivate the habit to meaningfully 
engage in a democratic life. Ann Sharp reconstructed and put forward the notion of community of 
inquiry as the guiding educational model for the philosophy for children movement (Gregory & 
Laverty, 2017).  

 
Here is an illustration of a classic example of a Mendham P4C session (noting that there are 

many other important forms of P4C and the ways of doing community of inquiry are always 
amendable): in a classroom of philosophy for children, students with diverse backgrounds and lived 
experiences gather together in a circle to read a selected text aloud. Normally each student reads a 
part of the text and everyone has a turn, so that they could share meaning with each other, read aloud 
with expression and emotions, and learn to carefully listen to others. Then, teachers collect students’ 
questions about issues they find puzzling, and write them down on the chalkboard/whiteboard for 
further discussion. After students choose a question as the target of discussion, they inquire with each 
other, which often involves making assertions that are supported with reasons, clarifying one’s 
position, and providing (counter) examples. The goal of such inquiry is to “form a judgement about 
the matter that is reasonable, meaningful and practicable as they can manage (Gregory & Laverty, 
2017).” 

 
However, given the divided political and racial climates in the current era of globalization, a 

community may not always be a birthplace of respect, diversity and mutual learning; it can also 
generate conflicts, discontent, feeling of divergences, bullying and exclusion. Thus, the idea of 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

15 
 

 

 

 

building ‘an intellectually safe community’ where all participants can be challenged in their 
worldviews but at the same time feel supported and safe is a timely response to such concerns in P4C 
(Butnor, 2012). 

Literature Review 

The radical nature of P4C in transforming our vision of the function of philosophy from a 
sphere for intellectual elites to a place for human beings with diverse age and life experiences 
generates skepticism and debate. The first question is whether children are intellectually mature 
enough for philosophy (Daniel & Auriac, 2011). This debate can be traced back to Plato and 
Aristotle’s views on the nature of children and their negative opinions on children’s intellectual 
ability and appropriateness to philosophize (Kennedy, 2006). In modern educational theories, 
according to Piaget, children are not equipped with the ability to do abstract thinking (Piaget, 1931). 
However, Vygotsky’s emphasis on the role of social interactions in cultivating children’s intelligence 
potential provides strong support for the idea that age is not the determining factor of children’s 
cognitive abilities (Roberts, 2016). Recently, cognitive scientists have shown that children have much 
higher level of cognitive abilities than Piaget estimated (Gopnik, 2009). Furthermore, philosopher 
Gareth Matthews stressed the freshness of children’s ideas that he discovered through his 
philosophical discussion with young people, and thus rejected deep-rooted condescending attitudes 
behind this ‘children are not capable of doing philosophy’ argument (Matthews, 1980, 1994).  

 
The second question concerns the evaluation of P4C, which is related to questions such as 

whether P4C is effective according to different metrics and if there are various ways this program can 
be implemented to benefit more students, especially those who are challenged and disadvantaged, at 
an affordable cost (Gorard, Siddiqui, & Huat See, 2015).  

 
Since the 1970s, the outcomes measured in P4C research can be divided into two categories: 

(1) cognitive outcomes (2) socio-psychological outcomes related to attitudes toward academics, 
prosocial attitudes and behavior. Even though there are emerging studies that appraise the 
effectiveness of P4C in the socio-psychological field (Abbasi & Ajam, 2016; Dasí, Quintanilla, & 
Daniel, 2013; Scholes et al., 2016), the extant literature is still limited. Most studies focus on the goal 
of the Philosophy for Children program to provide a more formal training to develop students’ 
cognitive outcomes. This includes direct assessment of students' reasoning skills and abilities, 
comprehensive skills, and academic performance (García-Moriyón, Rebollo, & Colom, 2005; Gorard 
et al., 2015; Gregory, 2011; Säre et al., 2016; Trickey & Topping, 2004) as indicators to students' 
learning progresses.  

 
Previous Review of the Evaluation of P4C 

 
P4C may have a positive effect on students’ cognitive abilities. In 2004 and 2005, two 

systematic reviews (García-Moriyón et al., 2005; Trickey & Topping, 2004) were conducted to 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

16 
 

 

 

 

synthesize research on the effectiveness of P4C. First, the quantitative systematic analysis by Trickey 
and Topping (2004) investigated the influence of P4C on students in general, with the conclusion 
that P4C has a moderate positive effect on students’ abilities with low variance. It collected eight 
controlled experiments regarding Philosophy for Children from 1970s to 2002. 

 
Even though the relationship between the two has not been yet been accepted by researchers 

(Hidi, Renninger, & Krapp, 2004), they combined the cognitive outcomes and affective abilities 
without a theoretical foundation for doing so.  

 
The second study is a meta-analysis conducted by García-Moriyón, Rebollo, and Colom 

(2005), in which they examined the relationship between P4C and reasoning skills as outlined in 18 
studies published from 1976 to 2002, with the finding that P4C has a positive moderate influence on 
students’ reasoning abilities. This meta-analysis included 18 experiments. The results showed 
significant differences among post-test experiments, single group studies with pre and post-test, and 
controlled experiments, in which the more rigorous controlled experiments tended to show lower 
effect sizes.  

 
These two reviews (García-Moriyón et al., 2005; Trickey & Topping, 2004) provided 

significant contributions in understanding the impacts of implementing Philosophy for Children in 
K-12 education. However, a new meta-analysis is needed to address the following issues in the 
contemporary situation: First, after the publication of the two earlier meta-analyses, a larger collection 
of literature on the effects of P4C on cognitive outcomes has been generated with an increasing rigor 
of study designs, a larger number of participants, and follow-up studies (Fair et al., 2015a; Fair et al., 
2015b; Topping & Trickey, 2007a, 2007b). Thus, researchers now have the opportunity to improve 
the rigor of a systematic analysis by only including studies with random controlled experiments or 
quasi-experiments and analyze these findings in detail through moderator analyses to find if the 
relationship between cognitive outcomes and P4C intervention is depended upon other variables 
such as duration of the program, sample size, et cetera. Second, since the P4C movement has spread 
worldwide and research was conducted on different continents (Lam, 2012; Marashi, 2008; Nia, 
2014; Youssef, 2014), a meta-analysis at this stage can involve an exhaustive search globally in English 
and capture the multiplicity of P4C practices worldwide. Thus, the present study aims at conducting a 
more recent and detailed analysis of the literature to help educators acquire a clearer understanding 
of the effectiveness of Philosophy for Children movement as a globalized phenomenon. 

 
The Present Study 

 
The purpose of the current meta-analysis is to examine the reported effectiveness of P4C from 

2002 to 2016, immediately following the publication of the two articles that analyzed studies from 
1970s to 2002. In addition, this meta-analysis examines which variables –participant age, 
socioeconomic status, study location, assessment measure, duration– of the intervention might 
moderate the magnitude of the aggregated effect sizes.   



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

17 
 

 

 

 

 
Through this meta-analysis, the researchers hope to find answers for the following question: 
 
1. What does the cumulative research suggest regarding the overall effectiveness of P4C on 

students’ cognitive abilities? 
2. Do study design, students' backgrounds (grade level and socio-economic status), location    

and duration of intervention, and characteristics of cognitive outcome measurements 
influence the magnitude of the effect size of included studies? 

Methodology 

In this study, the effectiveness of P4C was tested through a meta-analysis, which is a method 
that merges the results of many independent researchers, conducted on a particular topic and 
performs statistical analysis (Çoğaltay & Karadağ, 2016).  

Study Search and Retrieval 
 

This study included the online databases British Education Index, ERIC, Education Full Text 
(H.W. Willson), Education Source, Academic Search Complete, PsycINFO database from 2002 to 
2016. The keyword used was philosophy n2 children, which means it specified 2 maximum intervening 
words between philosophy and children, in any order. The researchers included both referred 
published journals and doctoral dissertations. Second, the researchers conducted a non-electronic 
journal search. The index of the journal Thinking: Philosophy for Children was consulted for articles. 
Then, potential relevant articles were retrieved from a library. The third was a google scholar search 
engine, Journal of Philosophy in Schools, as well as references listed in collected studies. Through the 
initial searches, 1180 articles were potentially relevant.  

Inclusion Criteria 

 
In order to be included in this meta-analysis, studies had to meet the following criteria: 
 

1. Participants: The population of interest was pre-collegiate students enrolled in a Philosophy 
for Children program and their control-group counterparts. College studies and teacher 
education research were excluded from the study.  

2. Intervention: Philosophy for Children has various names and diverse ways of practicing in the 
world. In this meta-analysis, we included studies that are under the names of P4C, Philosophy 
for Children, Philosophy with Children, and PwC. All the included studies must have explicit 
pedagogical markers of "community of (philosophical) inquiry" that shares the common 
practices of providing stimulus (stories, questions, pictures, or other media), students' 
questioning, and building on each other’s ideas.  



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

18 
 

 

 

 

3. Publication: The retrieved study should be published between 2002 and 2016 in a refereed 
journal or as a thesis/dissertation.  

4. Research design: 1) The study must be either random controlled experiments or quasi-
experiments. 2) A quantitative measure of outcomes was used in the study to calculate the 
effect sizes of the intervention. 3) The outcome variables contained a measurement of 
cognitive outcomes, such as reasoning ability, comprehension ability, general cognitive ability, 
and academic development. 4) This meta-analysis focused on comparing the cognitive 
outcomes of P4C as the experimental group with other control groups where participants did 
not receive any thinking skill intervention. Thus, studies that did not contain a control group 
were excluded.  
 

    To ensure all studies included were well-designed and able to provide enough data for the 
computation of effect sizes, researchers left out studies that failed to conform to any of those criteria. 
Thus, a considerable number of studies were excluded in this stage particularly because many of them 
adopted qualitative methodology, which could not provide enough effect size for meta-analysis. This is 
because of the nature and limitation of meta-analysis itself as it applies only to research studies that 
produce quantitative findings. That is, studies using quantitative measurement of variables and 
reporting descriptive or inferential statistics to summarize the resulting data (Lipsey & Wilson, 2001). 
This process ruled out qualitative forms of research such as case studies, ethnography, and 
‘naturalistic’ inquiry (Lipsey & Wilson, 2001). 

 
    Because of the large amount of literature, there were two steps of screening during the selection 

of included studies. First, one of the authors screened the titles and abstract of each of the 1,180 
studies. 44 articles which met all the criteria of inclusion remained from the initial screening. Then, 
the full text of each of the 44 studies was retrieved and scrutinized. During this second screening of 
44 full texts, 28 more studies failed to meet the requirements of the inclusion criteria. Finally, the 
remaining 16 studies were subject to the coding process.  

 
Coding Procedure 

 
The coding process is a data extraction process, picking clear and appropriate data from the 

pile of complex information (Çoğaltay & Karadağ, 2016). The manual for coding studies was 
developed by the researchers before proceeding to the coding.  

 
Content validity for the coding sheet and coding manual was determined originally by 

submitting to scholars/researchers for feedback on the appropriateness of variables and categories 
created in this study. The first scholar works primarily on culture and curriculum, and the second 
professor is in educational psychology. There were two coders in this study. The first coder is the first 
author of this meta-analysis; the second coder is a doctoral student and the 3rd co-author. Both of the 
two coders have received statistical education for quantitative research.  

 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

19 
 

 

 

 

After the coding manual was created, two coders initially met to go over the coding manual 
until everything was clear. The coders scrutinized each of the articles and extracted the variables and 
outcomes from the studies and input them into an excel document. To determine interrater 
reliability, the two researchers independently coded five studies (31.25% of the 16 articles) to ensure 
that the inclusion/exclusion criteria were met. The researchers achieved an interrater reliability of 
90.0% across those studies. Analysis of coder disagreements resulted in the refinement of some 
definitions and decision rules for some codes. Then, each coder individually coded the remainder of 
the studies.  

 
During the coding process, the first coder contacted the original authors from two different 

references for standard deviations and means to calculate the effect sizes. One set of data was 
obtained and another contact for data was not successful.  

 
Data Analysis 

 
Effect size computation, test of homogeneity and moderator analysis were conducted in the 

stage of data analysis. The effect size acquired in the meta-analysis study is a standard measure value 
used to determine the strength and direction (Çoğaltay & Karadağ, 2016) of the effectiveness of 
Philosophy for Children program on students’ cognitive outcomes. Cohen’s d was used to adjust and 
determine the effect sizes of each study. All the effect sizes in each study were aggregated to one effect 
size as the cognitive outcome. In meta-analysis, the unit of analysis is the individual research study, 
and any two or more effect sizes that come from the same study are statistically dependent (Lipsey & 
Wilson, 2001). Furthermore, all data analyses involving effect sizes were weighted analysis. 

 
Two main models, namely fixed effects model and random effects model, were utilized in the 

analysis of heterogeneous distribution of effect sizes. Under a fixed effects model, an effect size 
observed in a study is assumed to estimate the corresponding population effect with random error 
that stems only from the chance factors associated with subject-level sampling errors in that study 
(Lipsey and Wilson (2001). If it is believed that the research is not equal in terms of functionality, and 
if generalizations through the estimated effect size are to be made for greater populations, then the 
model that should be used is the random effects model (Çoğaltay & Karadağ, 2016).  

 
Last, the authors evaluated the data generated by the analysis and determined 1) whether P4C 

had a positive impact on students' cognitive abilities to address the first research question and 2) if 
characteristics of study design, students' background location and duration of intervention, 
characteristics of cognitive outcome measurements could be the moderator(s) of P4C on the effect 
size of cognitive abilities to answer the second research question.   

 
 
 
 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

20 
 

 

 

 

Results 
 

Six studies were excluded from the 16 research papers during the last stage of this analysis. 
One study was excluded because participants were younger than first grade (Säre et al., 2016) and this 
study only focuses on K-12 students. The study conducted by Othman and Hashim (2006) was 
excluded because the experiment’s control group was still participating in another thinking skills 
intervention. This study compared P4C to another thinking program (the Reader Response Program). 
Thus, the control group is not neutral. The control groups in all the included studies were not under 
any thinking skills intervention. Two studies were not included due to the lack of means and 
standard deviations to calculate the effect sizes (Colom, Moriyon, Magro, & Morilla, 2014; Walker, 
Wartenberg, & Winner, 2013). Another one (Gorard et al., 2015) was not from a peer-reviewed 
journal and thus excluded from the study. One study was excluded because its outcome measure was 
spiritual development, which is not considered to be within the realm of cognitive outcomes 
(Abaspour, Nowrozi, & Latifi, 2015).  

 
Studies Included 

 
A total of 10 controlled experiments were included in this analysis, which together report the 

findings of eight independent studies and two follow-up studies. Among the ten studies, nine were 
articles from peer-reviewed journals, and one is a dissertation (Youssef, 2014). Table 1 provides an 
overview of the characteristics of each citation included in the synthesis. The sample sizes in these 
studies ranged from 28 to 540, representing 1,509 students from second grade in elementary to 9th 
graders in high schools. The sample sizes of studies were adjusted in following way:  

 
First, if the sample size of one study in post-test was smaller than the pre-test due to the loss of 

participants, then the whole sample size of this study was coded according to the number of 
participants of the post-test. Second, in the case of one study with a follow-up study (Fair et al., 2015a; 
Fair et al., 2015b), the sample sizes of the independent studies were adjusted to the corresponding 
student groups with the follow-up studies. Thus, the overall sample sizes in this meta-analysis are 
smaller than those in the original literature.  

 
Regarding outcomes measures, four studies (Fair et al., 2015a; Fair et al., 2015b; Topping & 

Trickey, 2007a; Topping & Trickey, 2007b) used the Cognitive Ability Test, a standardized test called 
CAT in America and CogAT in United Kingdom. It measures students' verbal, Quantitative and 
nonverbal reasoning abilities (Lohman et al., 2001). Two studies (Lam, 2012; Marashi, 2008) chose 
the The New Jersey Test of Reasoning Skills (NJTRS), which was specifically designed to measure 
reasoning skills. One study (Naderi, 2014) selected Abedi's Test of Creativity which was formulated to 
measure rate of creativity based on Torrance Tests of Creative Thinking (TTCT). Another study 
(Youssef, 2014) utilized a standardized test called The Test of Reading Comprehension (TORCH) to 
measure reading comprehension ability of the students. One study (Abbasi & Ajam, 2016) developed 
a questionnaire to test the educational progress of science lessons. To verify the validity of the 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

21 
 

 

 

 

measurement, the researchers included 20 in-service teachers of second grade to review the 
instrument. In the study by Tok & Mazı (2015), both the Reading Comprehension Test and 
Listening Comprehension Test were designed by the researchers. They were developed through the 
framework of predicted objectives for the reading and listening comprehension learning field in the 
elementary fifth grade Turkish Course Curriculum. 

 
Overall Effectiveness of P4C 

 
The overall effect size aggregated from the ten studies was 0.43 with a 95% confidence 

interval, ranging from 0.33 to 0.53. According to Cohen’s Rule of Thumb (VanVoorhis & Morgan, 
2007), the mean effect size represents that P4C has a moderate, positive overall cognitive effect for 
students who are in 2nd to 10th grade.  

 
 In this study, the homogeneity test was found to be statistically significant (Q = 26.59, p < 
0.01), which means that there is more variability in effect sizes than would be expected from sampling 
error around the mean. Table 2 provides the overall results and omnibus test of this meta-analysis.   
 

Results of Moderator Analysis 
 

Since the homogeneity test was found to be statistically significant, a moderator analysis was 
used to find out the potential explanations for variance among effect sizes. In this meta-analysis, 
subgroup analysis was employed to detect moderating effects. Seven moderator variables were tested: 
grade level, socioeconomic status of students, location of studies, study design (random or quasi-
experiments), total time of intervention, outcome measures, and type of outcomes. Table 3 provides a 
detailed statistical description of the result of moderator analysis.  

 
In this study, two of the seven moderators revealed statistically significant effects. They were 

research location (two subgroups: Asia and western countries) and type of outcomes (three subgroups: 
general cognitive ability, reasoning skills and academic achievement). The tests of homogeneity 
indicated no statistical differences by grade levels, socio-economic status of participants, methods of 
group assignments, duration of the intervention, and outcome measures. The following is the detailed 
description of each subgroup analysis. 

 
Grade Level. The included studies were divided into two categories in terms of the grade 

levels: 2 to 5 (k=4) and 6 to 10 (k=6). As seen in table 3, the aggregated effect size (Cohen’s d) of 
studies which recruited grade 2 to 5 students was 0.51, and the average effect size of studies with 
grade 6 to 10 students was 0.42. QB was 0.75 (p > .75). From the results of this moderator analysis, no 
significant difference was found between effect sizes of studies according to the grade levels of the 
samples.  
 
 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

22 
 

 

 

 

 
Table 1  

Characteristics of Included Studies 

Reference Study Type Location Sample 

Size 

Grade/Age 

Level 

Outcome Measure Effect Size Variance 

(Abbasi & 

Ajam, 2016) 

Intervention Iran 50 Second Questionnaire of 

Educational 

Progress* 

0.870 0.080 

(Fair et al., 

2015b) 

Intervention United States 177 Seventh CogAT 0.590 0.020 

(Fair et al., 

2015a) 

Follow-Up United States 115 Seventh Grade- 

Two Years after 

CogAT 0.570 0.030 

(Lam, 2012) Intervention China 28 Secondary 

School First 

Grade 

NJTRS 0.590 0.190 

(Marashi, 

2008) 

Intervention Iran 60 Eighth NJTRS 1.100 0.070 

(Naderi, 

2014) 

Intervention Iran 60 High School 

First Grade 

Abedi's Test of 

Creativity 

1.190 0.070 

(Tok & Mazı, 

2015) 

Intervention Turkey 74 Fifth Grade Reading 

Comprehension 

Test* and Listening 

Comprehension 

Test* 

0.162 0.035 

(Topping & 

Trickey, 

2007a) 

Intervention United 

Kingdom 

540 Ten-year-old 

students 

CAT 0.25 0.01 

(Topping & 

Trickey, 

2007b) 

Follow-Up United 

Kingdom 

183 Ten-Year-Old 

Students (Two 

Years After) 

CAT 0.400 0.020 

(Youssef, 

2014) 

Intervention Australia 222 Sixth Grade Reading 

Comprehension Test 

0.340 0.020 

Note: CogAT: Cognitive Ability Test (American Version); CAT: Cognitive Ability Test (United Kingdom Version); 

          NJTRS: New Jersey Test of Reasoning Skills; *: Tests developed by researchers 

 
 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

23 
 

 

 

 

Table 2  

Overall Results and Omnibus Test of P4C Studies 

 k N Median ES 
(d) 

Fixed Effect Random Effect Q 

ES (d) 95% CI ES (d) 95% CI 

P4C 10 1,509 .58 .43 [.33, .53] .50 [.33, .66] 26.59** 

Note: k = study size; N = total number of participants; CI = confidence interval; Q = omnibus test of homogeneity.    

          ** p ＜ .01 
 

 
Socio-Economic Status of Students. In this sample of studies, we used two categories for the 

socio-economic status (SES) of the participants. The first group consisted of students who received 
free lunch, or were classified as ‘economically disadvantaged’ by the local districts. The second group 
of students are not identified as part of the free-lunch program, or was classified as from middle (or 
upper) class families. However, no significant heterogeneity in effect sizes was found between the two 
groups of students. 
 

Research Location. This meta-analysis covers 5 studies conducted in the Western world and 5 
in Asian countries. Five of the examined studies conducted in Asian countries: Iran, Turkey, and 
China. The other five studies come from Western countries: United Kingdom, Australia and the 
United States.  
 

The first reason for doing this moderator analysis is because P4C as an educational movement 
has its roots in the Western philosophical traditions, which bring about the authors’ uncertainty of 
the viability of its globalization. For example, P4C’s exclusive emphasis on dialogue (Gregory, 2011), 
which is different from some non-Western philosophical practices such as contemplation, may 
negatively impact its implementation and effectiveness.  

 
Another impetus for this analysis is that of the socio-political environments in these non-

Western countries. For example, countries such as Iran and China might be more resistant to 
Western rhetoric and democratic schooling as a whole. In Is Respecting Children’s Rationality in Their 
Best Interest in an Authoritarian Context?, the authors (Ghazinejad & Ruitenberg, 2014) argued that 
P4C implementation in Iran must balance the teaching of critical thinking and the protection of 
children’s safety in their communities. Giving the consideration that P4C and its democratic 
educational ideals not only have conflicts with the extant educational systems but also may bring 
clashes between individual students and their communities, the authors wonder if the effectiveness of 
P4C will be influenced by these factors at all. 

 
Through moderator analysis, a significant difference between the two groups was found (Q = 

5.16, p < .05). The studies in Asian countries had higher effect sizes (d=0.69) than those studies 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

24 
 

 

 

 

conducted in Western countries (d=0.39). Yet, the five studies in Asian countries had significantly 
lower sample size (n=272) than the five studies in non-Asian countries (n=1237). 

 
Study Design. To warrant the rigor of this meta-analysis, the authors set up stringent criteria 

for the inclusion of studies in which only random controlled trials and quasi-experiments were 
brought in the synthesis. From the moderator analysis, no significant difference was found between 
effect sizes of random controlled experiments and quasi-experiments which were included in this 
meta-analysis. 

 
Duration. The authors of this study divided the literature into three subgroups based on the 

duration of interventions: 5 to 20 hours (k=4), 21 to 40 hours (k=3), and more than 40 hours (k=3). 
The result showed that none of the duration levels statistically varied from one another. Thus, there 
was no noteworthy difference between different levels of duration of intervention in the effectiveness 
of P4C on students’ cognitive outcomes.  

 
Outcome Measure: CAT or Non-CAT. Studies included were examined according to their 

outcome measures. Four studies using Cognitive Ability Tests were accepted as CAT subgroup; six 
studies applying other types of outcome measures were accepted as Non-CAT subgroup. No 
significant heterogeneity was found between these two subgroups.  

 
 Types of Outcomes. A significant difference among different types of outcomes was found 
(QB = 15.44, p < .001). The studies (Lam, 2012; Marashi, 2008) which tested the improvement of 
reasoning skills through P4C yielded the largest estimations (d=1.06), which is a large effect size. P4C 
used in improving general cognitive abilities (d=0.40), which is a moderate effect size. Reading 
comprehension ability (d=0.28) is a small effect size. This suggests that P4C has significant, positive 
influence on students’ reasoning skills, and moderate effects on general cognitive ability and 
comprehension ability.  
 

Summary 
 

The first research question in this meta-analysis concerned the direction and magnitude of the 
effectiveness of P4C on students’ cognitive ability. The studies analyzed here showed an overall 
positive medium effect size on cognitive outcomes in general.  

 
The second question was whether and how the effectiveness of P4C differed significantly 

depending on the moderator variables. The moderator analysis found statistically significant results in 
regard to the study location and outcome types of these studies. No significant differences were found 
as to different grade levels, socio-economic statuses of participants, methods of group assignment, 
duration times of intervention, and cognitive measures. The results suggest that Philosophy for 
Children has a positive moderate influence on students’ cognitive outcomes.  

 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

25 
 

 

 

 

Discussion 
 

Ten studies were included in this meta-analysis to determine the effects of Philosophy for 
Children program on students’ cognitive abilities, and what characteristics of the intervention, 
students and outcomes measures could influence the magnitude of such effect.  

 
The Overall Effectiveness of P4C 

 
According to the findings of this meta-analysis, the Philosophy for Children program has 

shown a moderate, positive influence on students’ cognitive outcomes. This result corroborates the 
previous literature on the program that states that P4C has a positive impact on students’ various 
types of cognitive abilities (Fair et al., 2015a; Fair et al., 2015b; García-Moriyón et al., 2005; Topping 
& Trickey, 2007a, 2007b; Trickey & Topping, 2004). 

 
The cognitive outcomes comprise general cognitive ability, reasoning skills, creative thinking 

abilities, educational progress in science, reading and listening comprehension abilities. Among all of 
these types of cognitive outcomes, the Philosophy for Children program has large aggregated positive 
effect on students’ reasoning skills, while moderate influences on other cognitive domains. The 
previous P4C meta-analysis that focused on reasoning abilities (García-Moriyón et al., 2005) also 
indicated the positive impact of P4C on students’ reasoning skills. However, as the number of studies 
selected in this meta-analytic review was limited, the interpretation of those aggregated results needs 
to be cautious. 

 
Discussions about Findings between P4C and Students’ Grade Levels 

 
As stated in the results section, there was no statistically significant cognitive outcome in the 

effectiveness of P4C based on the grade levels of students. This result sheds lights on the question 
regarding P4C and students’ age. Philosophy education is traditionally assumed to be appropriate for 
students no younger than secondary school age (Lipman & Sharp, 1978). But this moderator analysis 
indicates that both the studies with grade 2 to 5 students and the studies with grade 6 to 10 students 
benefited from P4C program (grade 2-5: d=0.51; grade 6-10: d=0.42). There was no statistically 
meaningful difference between the aggregated effect sizes of the two subgroups.  

 
 

 
 
 
 
 
 
 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

26 
 

 

 

 

Table 3  

Moderator Testing of Study 

         Variable k     N   d 95% CI     QB ANOVA 

Research Location       

         Asia 5 272 .69 [.46, .91]  >N-A   

         Non-Asian Countries 5 1,237 .39 [.27, .51]                           5.16*  

Grade at Intervention       

         2-5 4 416 .51 [.34, .69]   

         6-10 6 1,093 .42 [.29, .55]  0.75  

SES of Participants       

         Disadvantaged 4 1,015 .40 [.27, .53]   

         Others 6 494  .55 [.37, .72]                           1.74  

Methods of Group Assignment       

         Random 4 811 .44 [.33, .54]   

         Quasi Experiment 6 698 .52 [.38, .66]                           0.83  

Total Time of Intervention       

         5-20 Hours 4 445 .34 [.18, .51]                                                          

       21-30 Hours 3 579              .28 [.13, .43]   

       More than 40 Hours 3 427                    .47               [.28, .66]  2.41  

Outcome Measure       

       CAT or CogAT 4 1,015 .40 [.27, .53]   

       Others 6 494 .55 [.37, .72]  1.74  

Type of Outcomes       

       General Cognitive Outcomes 4 1,015 .40 [.27, .53]   

       Reasoning Skills 2 148 1.06 [.72, 1.40]  > C & R 

       Reading Comprehension 2 296 .28 [.06, .50]                      15.44***  

Note: k = study size; N = number of participants; CI = confidence interval; QB = between-groups test of homogeneity;  

ANOVA = significant result. * p ＜ .05, *** p ＜ .0 

 

 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

27 
 

 

 

 

Discussions about Findings between P4C and Locations 
 

 Another significant finding of this study is where the research was conducted could 
determine its effect sizes. A statistically significant difference was found between the effect sizes of 
studies in Western (d = 0.39) and non-Western (d=0.59) countries. From this result, it seems that the 
globalizing Philosophy for Children program has generated more positive influences on students’ 
cognitive outcomes in non-Western countries than Western countries. This is not expected by the 
authors in that considering P4C’s Western philosophical heritage, this program might not be suitable 
to the socio-political and philosophical contexts of non-Western educational settings and thus has less 
positive outcomes. There are several possible accounts for this phenomenon.  

 
First, the studies in Asia have smaller sample sizes. Because P4C is still new to educators and 

researchers in those countries (Lam, 2012; Marashi, 2008), including Iran, China and Turkey, these 
studies are often pilot studies with small sample sizes. Moreover, since P4C was initiated in the 
United States in the 1970s (Brandt, 1988), it is more relatively well-known to the educators in the 
United States, United Kingdom, Australia and other Western countries. Thus, studies conducted in 
these areas tended to evaluate P4C in large school districts (Fair et al., 2015b; Toppings & Trickey, 
2007a; Youssef, 2014). In this meta-analysis, the mean sample size of Western studies is three times 
higher than the mean sample size of non-Western studies. Smaller sample sizes may contribute to the 
quality of teacher education and P4C implementation. Pedagogically speaking, all the studies have 
utilized community of inquiry as the core pedagogy. Though it might be the case that there are more 
experienced practitioners in countries where P4C is more well-known or in studies with larger scale, it 
is also possible that the reverse is true. This is because the practitioners in pilot studies may have 
received more focused teacher education while in studies with large sample size teacher education and 
motivation for practicing P4C are not easy to control.  

 
Another possible explanation is that several studies in non-Western countries tested the 

improvement of reasoning skills among students (Lam, 2012; Marashi, 2008; Othman & Hashim, 
2006), while no Western research included here specifically examined the reasoning abilities of 
students. According to the moderator analysis regarding the effect sizes of studies with different types 
of outcomes, there is a statistically significant difference between reasoning skills and other types of 
outcomes. If P4C is more effective to the improvement of reasoning skills, then the discrepancy 
between the effect sizes in Western and non-Western studies is understandable.  

 
Although none of the studies from non-western countries in this meta-analysis seemed to 

utilize or create P4C philosophical texts that catered to their specific philosophical, socio-political, 
and educational environments, it is worthwhile to consider the complex and nuanced cultural and 
political consequences of introducing P4C to non-Western countries from the lens of de-colonial 
theories and democratic education (Ghazinejad & Ruitenberg, 2014; Ndofirepi & Cross, 2015), and 
also the possibility of P4C being transformed by its globalization (Gregory, 2011).  

 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

28 
 

 

 

 

Discussions about Findings between P4C and Duration of Interventions 
 

The moderator analysis showed that the P4C’s influence was not moderated by the duration 
of the intervention. This was not expected since several studies (Fair et al., 2015a; Fair et al., 2015b; 
García-Moriyón et al., 2005; Topping & Trickey, 2007a) have proposed that P4C should be 
implemented for a significant period of time before the program shows results. That being said, our 
result resonates with one P4C study (Fair et al., 2015b). In this project, the authors replicated a 
previous experiment conducted by Topping and Trickey (Topping & Trickey, 2007a), in which they 
shortened the duration of the P4C intervention to less than half of the former one: from 58 weeks to 
22 to 26 weeks. The result showed that P4C still had a moderate effect on students’ general cognitive 
ability. This suggests that a short time of exposure to P4C may also have a meaningful impact on 
students’ cognitive outcomes.  

 
Discussions about the Excluded Literature 

 
A large number of studies were excluded in the process of the analysis. The first reason is 

because the majority of studies in the field of P4C are qualitative and theoretical, whereas the 
methodology employed in this study is a quantitative meta-analysis that needs to extract the data from 
many independent studies conducted on a particular topic and perform statistical analysis (Çoğaltay 
& Karadağ, 2016). For example, numerous insightful articles regarding P4C have been published in 
Africa (Di Masi & Santi, 2016; Ndofirepi & Cross, 2015), but none of them was quantitative and 
could be used in this study.  

 
The second reason is that even if some studies utilized a quantitative methodology, they often 

lacked sufficient information especially for the means and standard deviations for the researcher to 
compute effect sizes.  

 
Third, the result of exhaustive literature search and process of study inclusion/exclusion 

showed that more rigorous quantitative studies regarding P4C program are still needed. The 
researchers gathered over 1180 studies at first, after coding procedure, there were only 16 studies 
remained. Throughout the data analysis process, six more articles were excluded from the study. The 
main reason of this phenomenon is that the majority of the literature regarding P4C is qualitative 
and theoretical. Due to the nature of meta-analysis, which is a quantitative synthesis study, it cannot 
process and analyze qualitative and non-empirical literature (Lipsey & Wilson, 2001). In addition, 
there is not only few quantitative experiments in P4C, but the data produced by the studies are not 
sufficient enough for computing an effect size. Thus, this suggests that this field needs more studies to 
form a larger cluster of rigorous research. 

 
Moreover, some studies gave a novel practice and detailed observation of children who are 

below the age of five. To narrow down the age to grade 1-12 students, one study about P4C’s 
effectiveness on reasoning skills was excluded, but it definitely shows the potential of teaching and 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

29 
 

 

 

 

introducing philosophy to very young children. For example, the study conducted by Dasi et al. 
(2013) showed a clear significant improvement in socio-psychological abilities among the 5-year-old 
children and a partial improvement in the 4-year-old children after participating in a few sessions of 
the P4C program. Also, one study (Säre et al., 2016) showed that P4C positively influenced 
preschoolers’ verbal reasoning skills. These studies provide information for educators and researchers 
to understand the unfamiliar area in which young children are involved in rather than exclude them 
from this philosophy. Future research can consider examining how P4C affects the cognitive outcome 
of children in kindergarten or preschool. Types of P4C study outcomes can be expanded from 
cognitive outcomes to psychological or social outcomes.   

 
Conclusion and Suggestions for Future Research 

 
P4C was found to have a moderate, positive overall effect on students’ cognitive outcomes. 

The authors suggest that P4C may be considered as an effective thinking program for teachers in 
grade 2-10 education. Based on the findings of this meta-analysis, several recommendations and 
suggestions for future research are advanced: 

 
First, in addition to long-term implementation of P4C in classroom, a short time of exposure 

to P4C may also have a meaningful impact on students’ cognitive outcomes. The practice of P4C 
should not only be limited to the realm of long-term applications.  

 
Second, this study suggests that grade level is not a moderator of the effectiveness of P4C in 

improving students’ cognitive abilities. Moreover, a small number of studies (Dasí et al., 2013; Säre et 
al., 2016) have practiced P4C with very young children who are below the age of five. Thus, age 
should not be the sole reason for excluding students from philosophy education, and more studies 
are needed in terms of the impacts of P4C on very young children. As Lone and Burroughs have said 
(2016), at one time or another we all ask philosophical questions of some kind, consider our values 
and reflect on the rightness and wrongness of our actions. It is possible that all children, regardless of 
age and grade level, have the capacity and interest to engage in philosophical activities (Lipman, 
2009). 

 
Lipman and Sharp (1978) once questioned a presupposition of P4C which assumes 

philosophy education should be assigned to students who are either from gifted programs or from 
particular advantageous backgrounds. The results of this study also indicate that there is no 
statistically significant difference in the impact of P4C based on social-economic backgrounds. It is 
suggested that educators in P4C program should strive to build a community of inquiry that 
encourages students to share not only divergent social backgrounds and life experiences (Lipman, 
2009) but also different styles of thinking (Lipman & Sharp, 1978) so as to involve them in the 
classroom discussion.  

 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

30 
 

 

 

 

Different from previous meta-analyses, this study emphasized the exhaustive search for P4C 
studies around the world. The results show that during recent years, a considerable amount of 
practices have been taken in various continents. This study calls for more research and analyses that 
consider the nuances and details of P4C practices in different cultural, social, educational, linguistic, 
and philosophical contexts.  

 
Last, while P4C as a famous thinking skill program has been relatively examined (Daniel & 

Auriac, 2011; Fair et al., 2015a; Fair et al., 2015b; García-Moriyón et al., 2005; Säre et al., 2016; 
Topping & Trickey, 2007a, 2007b), limited research on socio-psychological outcomes have been 
generated in this field. More studies that explore the connection between community of inquiry, 
philosophical thinking and the socio-psychological development of children are strongly 
recommended.  

 

 

References 
Abaspour, N., Nowrozi, R. A., & Latifi, Z. (2015). Investigating the Effect of Educating Philosophy in 

the Children on the Spiritual Development of Female Students with 12-14 Years Old in the 
City of Isfahan. Journal of Education and Practice, 6(11), 162-166.  

Abbasi, Z., & Ajam, A. A. (2016). The Effects of Philosophical Stories on Emotional Intelligence and 
Educational Progress of Students in Science Lessons. Mediterranean Journal of Social Sciences, 
7(2), 282.  

Brandt, R. (1988). On Philosophy in the Curriculum: A Conversation with Matthew Lipman. 
Educational Leadership, 46(1), 34-37.  

Butnor, A. (2012). Critical Communities: Intellectual Safety and the Power of Disagreement. 
Educational Perspectives, 44, 29-31.  

Çoğaltay, N., & Karadağ, E. (2016). The Effect of Educational Leadership on Organizational 
Variables: A Meta–Analysis Study in the Sample of Turkey. Educational Sciences: Theory & 
Practice, 16(2).  

Colom, R., Moriyon, F. G., Magro, C., & Morilla, E. (2014). The Long-term Impact of Philosophy for 
Children: A Longitudinal Study (Preliminary Results). Analytic Teaching and Philosophical 
Praxis, 35(1).  

Daniel, M., & Auriac, E. (2011). Philosophy, Critical Thinking and Philosophy for Children. 
Educational Philosophy and Theory, 43(5), 415-435.  

Dasí, M. G., Quintanilla, L., & Daniel, M. F. (2013). Improving Emotion Comprehension and Social 
Skills in Early Childhood Through Philosophy for Children. Childhood & Philosophy, 9(17), 63-
89.  

Dewey, J. (1916). Democracy and Education: An Introduction to Philisophy of Education: Macmillan. 
Dewey, J. (1986). Experience and Education. Paper Presented at the The Educational Forum. 
Di Masi, D., & Santi, M. (2016). Learning Democratic Thinking: A Curriculum to Philosophy for 

Children as Citizens. Journal of Curriculum Studies, 48(1), 136-150.  



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

31 
 

 

 

 

Fair, F., Haas, L. E., Gardosik, C., Johnson, D., Price, D., & Leipnik, O. (2015a). Socrates in the 
Schools: Gains at Three-year Follow-up. Journal of Philosophy in Schools, 2(2).  

Fair, F., Haas, L. E., Gardosik, C., Johnson, D. D., Price, D. P., & Leipnik, O. (2015b). Socrates in 
the Schools From Scotland to Texas: Replicating a Study on the Effects of a Philosophy for 
Children Program. Journal of Philosophy in Schools, 2(1).  

García-Moriyón, F., Rebollo, I., & Colom, R. (2005). Evaluating Philosophy for Children. Thinking: 
The journal of philosophy for children, 17(4), 14-22.  

Ghazinejad, P., & Ruitenberg, C. (2014). Is respecting children's rationality in their best interest in an 
authoritarian context? Ethics and Education, 9(3), 317-328. 

Giménez-Dasí, M., Quintanilla, L., & Daniel, M.-F. (2013). Improving emotion comprehension and 
social skills in early childhood through philosophy for children. childhood & philosophy, 
9(17). 

Gopnik, A. (2009). The Philosophical Baby: What Children's Minds Tell us About Truth, Love & the 
Meaning of Life: Random House. 

Gorard, S., Siddiqui, N., & Huat See, B. (2015). Philosophy for Children: Evaluation Report and 
Executive Summary. Education Endowment Foundation, Millbank, UK.  

Gregory, M. (2011). Philosophy for Children and Its Critics: A Mendham Dialogue. Journal of 
Philosophy of Education, 45(2), 199-219.  

Gregory, M. R., & Laverty, M. J. (2017). In Community of Inquiry with Ann Margaret Sharp: 
Childhood, Philosophy and Education: Routledge. 

Hidi, S., Renninger, K. A., & Krapp, A. (2004). Interest, A Motivational Variable That Combines 
Affective and Cognitive Functioning. Motivation, Emotion, and Cognition: Integrative Perspectives 
on Intellectual Functioning and Development, 89-115.  

Jackson, T. (2001). The Art and Craft of ‘Gently Socratic’ Inquiry. Developing Minds: A Resource Book 
for Teaching Thinking, 3.  

Lam, C.-M. (2012). Continuing Lipman's and Sharp's Pioneering Work on Philosophy for Children: 
Using Harry to Foster Critical Thinking in Hong Kong Students. Educational Research and 
Evaluation, 18(2), 187-203.  

Lipman, M. (2004). Philosophy for Children’s Debt to Dewey. Critical and Creative Thinking, 12(1), 1-
8.  

Lipman, M. (2009). (USA) Philosophy for Children: Some Assumptions and Implications. Children 
Philosophize Worldwide: Theoretical and Practical Concepts, 9, 23.  

Lipman, M., & Sharp, A. M. (1978). Some Educational Presuppositions of Philosophy for Children. 
Oxford Review of Education, 4(1), 85-90.  

Lipsey, M. W., & Wilson, D. B. (2001). Practical Meta-analysis (Vol. 49): Sage Publications Thousand 
Oaks, CA. 

Lohman, D. F., Thorndike, R. L., Hagen, E. P., Smith, P., Fernandes, C., & Strand, S. (2001). 
Cognitive Abilities Test third edition. London: nferNelson. 
Lone, J. M., & Burroughs, M. D. (2016). Philosophy in Education: Questioning and Dialogue in Schools: 

Rowman & Littlefield. 



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

32 
 

 

 

 

Lone, J. M., & Green, M. (2013). Philosophy in High Schools: Guest Editors' Introduction to a 
Special Issue of Teaching Philosophy. Teaching Philosophy, 36(3), 213-215. 

Love, R. (2016). The Case for Philosophy For Children in the English Primary Curriculum. Analytic 
Teaching and Philosophical Praxis, 36(1).  

Marashi, S. M. (2008). Teaching Philosophy to Children: A New Experience in Iran. Analytic 
Teaching, 27(1), 12-15.  

Matthews, G. B. (1980). Philosophy and the Young Child: Harvard University Press. 
Matthews, G. B. (1994). The Philosophy of Childhood: Harvard University Press. 
Meskin, A., & Cook, R. T. (2012). The Art of Comics: A Philosophical Approach: John Wiley & 

Sons. 
Murris, K. (1992). Teaching philosophy with picture books: Infonet. 
Ndofirepi, A., & Cross, M. (2015). Child’s Voice, Child’s Right: Is Philosophy for Children in Africa 

the Answer? Interchange, 46(3), 225-238.  
Ndofirepi, A. P., & Shanyanana, R. N. (2016). Rethinking ukama in the context of ‘Philosophy for 

Children’in Africa. Research Papers in Education, 31(4), 428-441. 
Nia, A. T. (2014). Investigate the Effect the Philosophy for Children Program (p4c) on Reducing Trait 

Anger in Teens. Stud, 4(2), 449-455.  
Othman, M., & Hashim, R. (2006). Critical Thinking & Reading Skills. Thinking: The Journal of 

Philosophy for Children, 18(2), 26-34.  
Peirce, C. S. (1868). Some Consequences of Four Incapacities. The Journal of Speculative Philosophy, 

2(3), 140-157.  
Peirce, C. S., & Houser, N. (1998). The Essential Peirce: Selected Philosophical Writings (Vol. 2): Indiana 

University Press. 
Piaget, J. (1931). Children's Philosophies.  
Roberts, A. F. (2006). The effects of a teacher development programme based on Philosophy for 

Children. University of the Western Cape.    
Säre, E., Luik, P., & Tulviste, T. (2016). Improving Preschoolers’ Reasoning Skills Using the 

Philosophy for Children Programme. Trames: A Journal of the Humanities and Social Sciences, 
20(3), 273.  

Scholes, L., Lunn Brownlee, J., Walker, S., Johansson, E., Lawson, V., & Mascadri, J. (2016). 
Promoting Social Inclusion in the Early Years of Elementary School: A Focus on Children’s 
Epistemic Beliefs for Moral Reasoning. International Journal of Inclusive Education, 1-14.  

Sharp, A. M., Reed, R. F., & Lipman, M. (2010). Studies in Philosophy for Children: Harry Stottlemeier's 
Discovery: Temple University Press. 

Tok, Ş., & Mazı, A. (2015). The Effect of Stories for Thinking on Reading and Listening 
Comprehension: A Case Study in Turkey. Research in Education, 93(1), 1-18.  

Topping, K. J., & Trickey, S. (2007a). Collaborative Philosophical Enquiry for School Children: 
Cognitive Effects at 10–12 Years. British Journal of Educational Psychology, 77(2), 271-288.  

Topping, K. J., & Trickey, S. (2007b). Collaborative Philosophical Inquiry for Schoolchildren: 
Cognitive Gains at 2‐year Follow‐up. British Journal of Educational Psychology, 77(4), 787-796.  



ANALYTIC TEACHING AND PHILOSOPHICAL PRAXIS VOL. 39, ISSUE 1 (2018) 

  
 

33 
 

 

 

 

Trickey, S., & Topping, K. J. (2004). ‘Philosophy for Children’: A Systematic Review. Research Papers 
in Education, 19(3), 365-380.  

van der Straten Waillet, N., Roskam, I., & Possoz, C. (2015). On the Epistemological Features 
Promoted by ‘Philosophy for Children’and Their Psychological Advantages When 
Incorporated into RE. British Journal of Religious Education, 37(3), 273-292.  

Vansieleghem, N., & Kennedy, D. (2012). Philosophy for children in transition: problems and 
prospects (Vol. 15): John Wiley & Sons. 

VanVoorhis, C. W., & Morgan, B. L. (2007). Understanding Power and Rules of Thumb for 
Determining Sample Sizes. Tutorials in Quantitative Methods for Psychology, 3(2), 43-50.  

Walker, C. M., Wartenberg, T. E., & Winner, E. (2013). Engagement in Philosophical Dialogue 
Facilitates Children’s Reasoning about Subjectivity. Developmental Psychology, 49(7), 1338.  

Youssef, C. (2014). A Multilevel Investigation into the Effects of the Philosophical Community of 
Inquiry on 6th Grade Students' Reading Comprehension, Interest in Maths, Self-esteem, Pro-
social Behaviours and Emotional Well-being (Doctoral Dissertation). Queensland University of 
Technology, Australia.  

 
Address Correspondences to: 

Sijin Yan (Correspondence Author), Texas A&M University, Teaching, Learning & Culture 

Department, Philosophy Department. 321 YMCA Building, College Station, TX 77843-4237. 

Email: yansj101@tamu.edu. Tel: 979-676-7911 

 

Lynne Masel Walters, Texas A&M University, Teaching, Learning & Culture Department, 362 

Harrington Office Building, College Station, TX 77843 Email:  lynne-walters@tamu.edu Tel: 

979-845-8384 

 

Zhuoying Wang, Texas A&M University, Texas A&M University, Teaching, Learning & Culture 

Department, 111D Harrington Office Building, College Station, TX, 77843 Email: 

ustop2013wzy@tamu.edu Tel: 979-587-4603  

 

Chia-Chiang Wang, Department of Counseling, School, and Educational Psychology, University 

at Buffalo, 413 Baldy Hall, Buffalo, NY, 14260-1020 Email: chiachia@buffalo.edu Tel: 716-

645-1119 

mailto:yansj101@tamu.edu
mailto:lynne-walters@tamu.edu
mailto:ustop2013wzy@tamu.edu
mailto:chiachia@buffalo.edu

