Abstract Inspired by the UN Charter on the Rights of the Child, educational policy makers in Africa are requesting schools to implement democratic practices in the classroom. This study attempts to shed light on the effects of classroom democratic practices on cognitive and non-cognitive learning outcomes using cross-sectional data about student performance and indicators of classroom democracy from Francophone African countries. To simultaneously account for the cause and effect of democratic practices on learning achievements and the nested nature of the data, use is made of the error components-two stage least squares estimator. The results confirm the endogenous nature of learning outcomes and indicate that teachers and parents are the most important stakeholders in the process of having children internalize democracy in the classroom. It is recommended that teachers develop skills to assign and supervise pupils’ work in small groups in the classroom. Education stakeholders should take measures to preserve and promote indigenous cultural values because they facilitate learning achievements. Keywords: Classroom democracy, Educational outcomes Njong, A. M. Global Journal of Transformative Education (2020) Vol 2 DOI 10.14434/gjte.v2i1.31205 Open Access Published by the Global Insitutute of Transformative Education (http://www.gite.education) © Njong, A. M.,. 2020. Open Access This journal is distributed under the terms of the Creative Commons Attribution NonCommercial NonDerivative 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits unrestricted use, distribution, and reproduction without revision in any non-commercial medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, Classroom Democracy and Learning Outcomes in Primary Schools in Francophone Africa Aloysius Mom Njong Full listing of authors and contacts can be found at the end of this article. Introduction Francophone Africans are familiar with classrooms where the teacher is the principal actor with all atten- tion focused on them. Francophone Africa includes countries where French is the official language even if not all residents speak French, a result of the influence of historical French colonization. Like in many other countries, teachers transmit textbook knowledge to pupils who are encouraged to memorise the course material, perhaps only to reproduce it on demand, usually to pass an examination. In such traditional and authoritarian schools, textbooks and teachers are the main sources of knowledge, and children are hardly en- couraged to participate. There is a growing awareness that such teacher-centred and text-book centred in- structional approaches produce future passive citizens who fall short of critical thinking, decision-making and socialisation skills (Barron & Darling-Hammond, 2008; Nagaraju, Madhavaiah, & Peter, 2013). Given this aware- ness and inspired by the UN Charter on the Rights of the Child, educational policy makers in Africa are requesting the enhancement of democracy in the classroom which is more likely to promote children’s social, moral and academic development (UN General Assembly, 1989). Despite this policy directive, very lit- tle attention has been paid to the practice of democratic principles in the classroom and the delivery of edu- cation in a transformative manner. Perhaps research in this area has been hampered by the lack of readily available data, and difficulties in quantifying social education variables (Borghans, Duckworth, Heckman, & Ter Weel, 2008). The traditional education produc- tion function has centred on conventional schooling input-output variables perhaps because these are more easily measurable. Non-conventional inputs such as democratic practices have been neglected. As part of classroom democratic practices, I agree with Serpell & Marfo (2014), Bodrova & Leong (2007), and Mucherah & Mbogori (2019) who argue that learn- ing outcomes are influenced by the cultural values and norms of the local setting where children go to school. African countries have cultural values which are specif- ic and diverse and often different from those in devel- oped countries. Yet few attempts are made to explicitly 47 Njong, Classroom Democracy and Learning Outcomes factor in African cultural values in the education pro- duction processes. Thus, in this study an attempt is made to consider some aspect of African cultural identity as a special dimension of classroom democra- cy in the modelling, and to show how it contributes to the motivation of children’s educational achievements in Francophone Africa. In addition to issues of data availability, the focus of the study on Francophone African countries is to capture the hangovers of French colonialism on the school curriculum of these coun- tries. There may be some aspects of French colonialism and post-colonialism that interact with other factors to shape schooling realities in postcolonial Francophone African countries. As concerns learning outcomes, focus in the liter- ature has been on literacy and numeracy test scores (Glewwe & Jacoby, 1994; Glewwe, 2002; Glewwe & Kremer, 2006; Wamalwa & Burns, 2018). Non-cogni- tive outcomes have not been given the attention they deserve, perhaps because they are difficult to measure. The classroom should not only aim to produce individu- als who can read and write, but must go beyond to pre- pare active, sociable, culturally viable and well-behaved citizens of the country. Another motivation for this study is at the meth- odological level. From the foregoing, it is apparent that schooling produces several learning outcomes, such as formal cognitive skills and behavioural skills. These educational outcomes are jointly determined in the learning process and any single equation modelling may not capture the entire picture of interrelationships in the educational process. Therefore, I attempt to take into account the simultaneity and interdependence be- tween learning outcomes and inputs, leading to a more realistic formulation of the schooling production func- tion. Another important methodological issue is related to the nested nature of the data. Ignoring clustering in analysing hierarchically structured data will underesti- mate standard errors due to the decrease in the effec- tive sample size for explanatory variables defined at the cluster level (Baltagi, 2011). It is against this backdrop that this study seeks to establish the causal relationships between democracy in the classroom and learning outcomes in primary schools in Francophone African countries. Specifically, the study attempts to: i. analyse democratic practices in schools as an ap- proach to classroom management in Francophone African countries. ii. determine the effects of classroom democracy on learning outcomes in primary schools across Fran- cophone African countries. The study is organised as follows: Section 1 ex- plains the context and motivation of the study. Section 2 presents the conceptual framework wherein the concepts of classroom democracy and learning achieve- ments are defined and operationalized. Section 3 dwells on the methodology which describes the data used, model specification and estimation strategies. Section 4 is reserved for results and discussions. Section 5 con- cludes the study with a summary of the key findings, together with educational policy implications. Conceptual Framework In this section, I define the concepts of classroom democracy and learning outputs in the context of this study so as to establish the causal relations between them. Classroom Democracy A modern pedagogical approach to classroom democracy focuses on the promotion of self-learning (Fearnley-Sander, Moss, & Harbon, 2001). This means that pupils are actively engaged with their own learn- ing. To be more explicit, I consider classroom democra- cy as a set of classroom practices that put the children at the centre of the instructional process. This requires that each pupil is accepted as a unique personality and shown respect. It also requires that pupils’ opinions be considered in education decision-making. As members of the society make decisions in a democracy, so the pupils who are actual elements in the democratic class- room should actively participate in decision making. For example, the curriculum to be taught and rules to be obeyed at school and in the classroom should be de- termined together with pupils. Democracy comes along with freedom of thinking and expression. Consequently, pupils should be given the opportunity to express what they feel and think. The democratic classroom as a way of life is rein- forced by the UN Convention on the Rights of the Child (UN General Assembly, 1989) which proposes rela- tionship changes between adults and young children. Both parents and teachers are expected to be agents of change in the learning process of the child. In building a democratic classroom culture, the role of the teacher has to change from an authoritative personality to a fa- cilitator. The teacher is no longer that person dictating a lesson as pupils copy and assimilate, but is expected Global Journal of Transformative Education (2020) Vol 2 48 Njong, Classroom Democracy and Learning Outcomes to help pupils to cultivate the habit of accessing infor- mation by themselves and learning it (Maftoon & Shak- ouri, 2012). The teacher is expected to guide the pupils in their learning process, and not force them to learn. The parent is expected to promote processes designed to enable and empower children to learn conveniently. Such processes would involve providing children with basic needs such as adequate nutrition, health, and protection against abuse and violence. Another dimension of classroom democracy is enshrined in the concept of discipline in schools. It is well recognized that if discipline is not taken into consideration, the classroom/school environment will be unsafe and may disrupt the learning process as well as the academic achievement of the children (Mothata & Squelch, 1997; Levin & Nalon, 1991). Teachers and school administrators have a major role to play through guidance and counseling of children in order to instill in them a sense of responsibility and curb incidences of disruptive behaviour in schools. Every school has a code of conduct which every child is supposed to respect. When the rules are broken, it is expected that disciplinary measures be applied according to the edu- cational regulations in force. Therefore, it is important to factor in child discipline as a transformative educa- tional input that will lead to a healthy classroom envi- ronment in a democratic manner. Another dimension of classroom democracy may be borrowed from social cultural theory which holds that culture and social interactions are essential fac- tors in children’s acquisition of knowledge (Okonji, 1971; Bodrova & Leong, 2007; Marfo & Biersteker, 2011; Marfo, Pence, Levine, & Levine, 2011; Mucherah & Mbogori, 2019). The African child grows up in a local context where the mother tongue is spoken, traditional music is listened and danced to with formidable dexter- ity, domestic chores are often distributed among family members, and telling of folk stories are common prac- tices. Furthermore, the extended family system which includes, in addition to nuclear family; uncles, aunties, cousins and grandparents is order of the day (Degbey, 2012). Democracy as a way of living and everyday experiences should not be indifferent to these local contextual factors in which African children are grow- ing. Therefore, African socio-cultural values need to be extended to the classroom setting in order to properly motivate children in the learning process in the context. Global Journal of Transformative Education (2020) Vol 2 Learning Outcomes Learning outcomes are multidimensional and complex. However, to facilitate understanding I divide learning outputs into cognitive and non-cognitive com- ponents. According to Anderson et al (2001) the cognitive domain involves the acquisition of knowledge and the development of intellectual skills from the educational process. This way of learning encourages pupils to use their brains more effectively, and enables them to think and remember easily. This does not mean that memo- risation or repetition is encouraged, it simply ensures that the concepts learnt in class are understood. Out- comes from the cognitive learning process usually in- volve academic achievement tests in reading, language and mathematics. The non-cognitive component is made up of self-es- teem and sociability outcomes. According to Emler (2001), Olsen, Breckler, & Wiggens (2008) and Harter (2012), self-esteem is considered a child’s evaluation of self-worth and acceptance in the classroom. It is a judg- ment of the self. Research results indicate that there is a correlation between self-esteem and children’s devel- opment (Papadopoulos, Metsiou, & Agaliotis, 2011). For example, children with high self-esteem show greater motivation and take initiative. They are more prone to express sociable behavior and show more feelings of happiness (Baumeister, Campbell, Krueger, & Vohs, 2003). On the other hand, children who have low self- worth tend to treat themselves badly and are more likely to invite bad treatment from their peers (Saigal, Lambert, Russ, & Hoult, 2002). The consequences of low self-worth amongst pupils would include unhap- piness, symptoms of depression, and difficulties in forming and sustaining friendly relationships with their peers (Emler, 2001). The social outcome of the education process revolves around the social learning theory (Bandura, 1977). The theory posits that people can acquire new behaviours by observing and imitating the behavioural patterns of other people. The theory further posits that learning would also occur through practices of rewards and punishments. In this way, a behaviour that is often rewarded would persist, while the one that is constantly punished would most likely not be imitated. The classroom is a social setting, and pupils through their interactions with peers and teach- ers would pick up new patterns of behaviour that are likely to uphold the spirit of team work, cooperation and love for one another. 49 only the items that pass the reliability or internal consistency test in analysis improves measurment of the latent constructs and facilitate the convergence of estimations (Wooldridge, 2009). Based on the retained items in each construct I built composite indicators using multiple correspondence analyses (Njong & Nin- gaye, 2008). In this sub-section I analyse the measured latent variables/constructs of interest in this study. Indicators of classroom democracy The main explanatory variable in this study is class- room democracy. Drawing from the conceptualisation of classroom democracy in section 2, I build and anal- yse four composite indicators of classroom democracy. The first is the indicator of group work. This composite indicator is captured based on primary items shown on Table S1 (see Appendix). Table S1 indicates that about 38% of the teachers organised the children to work in small groups. About 67% of the teachers reported they lack the techniques to make pupils work as teams, while 56.3% of them complained that the classroom is inappropriate to organise children in smaller groups. The second construct of classroom democracy is an indicator of children’s rights. The primary items of children’s basic rights retained are displayed on Table S2. It shows that about 26.4% of the children attended nursery school, while 64% of them slept under mos- quito nets. As concerns health challenges, 17.7% of the children reported having hearing difficulties, and 23% had sight problems. Children nutrition seems to be a serious problem because about 57.5% of the children reported always feeling hungry in class. The third con- struct of classroom democracy is an indicator of child discipline. Child discipline was surveyed by getting responses to the statements captured on Table S3. It is observed on Table S3 that about 38% of the teachers punished or shouted at children in class. The method of punishment ranged from asking the pupils to copy exercises (32%), and /or do manual labour (49%). Note that about 5.5% of the teachers did administer corporal punishment although it is prohibited by law in these countries. The fourth construct of classroom democracy is an indicator of cultural identity. The statements that captured cultural identity, and based on which the cul- tural identity indicator is built, are displayed on Table S4. Table S4 shows that 22.4% of the children did not speak their mother tongue at home, while 69% of them did so occasionally. Only about 18.5% of the teachers used the native language spoken by the children of that Global Journal of Transformative Education (2020) Vol 2 Njong, Classroom Democracy and Learning Outcomes Methodology In order to understand the causal relationship between democratic classroom practices and learning achievement, secondary cross-sectional data collected by reviewing PASEC surveys were used. The PASEC survey methodology was a stratified samplying design where schools were first randomly selected, and grade 6 kids were subsequently randomly interviewed from each school. Given the stratified nature of the data collection process, multilevel analytical framework is adopted to better unravel the inter-relationships, but before addressing this, a discussion of the data genera- tion procedure is necessary. Data Since 1960 The Conference of Ministers of Educa- tion of Francophone countries, better known in French acronym as CONFEMEN1, has been making enormous attempts to promote the delivery of educational ser- vices and professional training (PASEC, 1998). To facil- itate this, it pilots a survey called ‘Programme d’Analyse des Systèmes Éducatifs de la CONFEMEN’ (PASEC) to collect data that are used to assess the efficiency of the educational systems of its member countries. The data I use in this study come from the PASEC survey conducted across 10 Francophone African countries in 2014 (PASEC 2017). As part of the PASEC survey design information was collected from pupils in grade 6 who are at the end of the primary school curriculum. The in- formation collected from these pupils included reading and numeracy test scores and the pupils’ characteris- tics. Information was also collected from the children’s parents, teachers and head teachers of the various schools relating to knowledge, welfare, pedagogic re- sources and governance issues. The 2014 PASEC design surveyed 31213 pupils in grade 6 nested in 671 schools across 10 countries2. The unit of analysis is the pupil. Measurement of Constructs Apart from the cognitive skills’ construct, the rest of the constructs in the survey data are captured using several items. I subjected the items in each construct to an internal consistency test and retained for analysis only those items with positive inter-items correlations and/or that loaded high on the construct in question. The Cronbach’s alpha test was used for 3-4 points Likert scales items, while the Guttman’s split lambda coefficient was used for yes/no responses. Retaining 1CONFEMEN is Conférence des Ministres de l’Éducation des États et Gouvernements de la Francophonie. 2The ten countries are: Benin; Burkina Faso; Burundi; Cameroon; Congo; Cote D’Ivoire; Niger; Senegal; Tchad; and Togo 3In the case of items with mixed scales, I used the Cronbach’s alpha test. Njong, Classroom Democracy and Learning Outcomes 50 locality. Above 70% of the teachers did not allow chil- dren talk or sing songs in their mother tongue in class. Table S4 also indicates that 77- 84% of the schools did not have textbooks in the language of the majority of the children. Dependent variables The dependent variables are learning outcomes which are cognitive and non-cognitive and are consid- ered endogenous, because they are determined with- in the system4. The cognitive outcomes are standard scores in reading skills and mathematics skills. This is a simple scale of test scores that are continuous from grade 2 through grade 6. The standard scores reported for each of the grade levels ranges from 100 to 400. I compute the mean value of the test scores across the grades for reading and math. These scores indicate that the higher the score, the higher the academic achieve- ment of the child. The non-cognitive outcomes are self-esteem and so- ciability. The items used to measure non-cognitive skills include measures of friendliness, happiness, self-worth, cooperation with one another, and feeling of security. Each child was asked to respond to the descriptive statements measuring self-esteem and sociability cap- tured on a 4-point Likert scale as follows: 1) strongly agree, 2) agree, 3) disagree, and 4) strongly disagree. The number of items in the survey that are related to self-esteem and sociability are ten and six respectively. Based on Cronbach’s alpha analysis, I selected seven and four items that loaded high on self-esteem and sociability respectively (see Tables S5 and S6). Table S5 shows children’s responses to self-esteem items. It shows that over 83% of the children reported perform- ing well in class. Approximately 81% of the children reported hard work to account for their good achieve- ment, while 74% of them attributed the good scores to the fact that class lessons are easy. Children reported having low scores for various reasons, for example about 47% of them don’t study enough, and 36% of them consider they are not intelligent enough, while 25% said the teacher insults them. Table S6 shows children’s responses to sociability items, where over 80% of them agreed having many friends in school. About 87% of the children were happy being in school because their peers like them. It is important to note that the learning outcome/ endogenous variables are measured in different units. The non-cognitive skills are measured using Likert scales in the survey dataset while the cognitive skills are mean test scores. Thus, in order to facilitate in- terpretation of changes on learning outcomes I trans- formed them into a common unit of measurement using standard z-scores. Model Specification To determine the effects of classroom democracy on learning outcomes an econometric model is spec- ified. Since the schooling inputs jointly produce the educational outcomes, a simultaneous equation mod- elling approach is deemed appropriate for this study. Simultaneous-equation models specify relationships in a system of two or more equations where the depen- dent variables display an interdependent relationship (Wooldridge, 2009; Greene, 2011; Baltagi, 2011). In a non-recursive model of this nature, the focus is on the relationship among the jointly determined dependent variables in the systems, which are considered endog- enous. I argue that these endogenous variables are considered cause and effect in the schooling production process. There is simultaneity or reciprocal causation among the dependent variables. The modelling sug- gests that certain outcomes could be affected directly by other outcomes of the educational process. What this means is that all the dependent variables are ex- plicitly taken to be endogenous to the system (Baltagi, 2011). That is, the endogenous explanatory variables are dependent variables from other equations in the system. Given the above, I can specify the following non-recursive system of simultaneous equations: Global Journal of Transformative Education (2020) Vol 2 4This is explained in greater detail in Methodology section. MTSijc=α0+α1ICDijc+α2RTSijc+α3ISOijc+α4ISEijc γXijc+ε1,ijc (1) RTSijc=β0+β1ICDijc+β2MTSijc+β3ISOijc+β4ISEijc+θWijc+ε2,ijc (2) ISOijc=δ0+δ1ICDijc+δ2MTSijc+δ3RTSijc+δ4ISEijc+ Gijc+ε3,ijc (3) ISEijc=ω0+ω1ICDijc+ω2MTSijc+ω3RTSijc+ω4ISOijc+πZijc+ε4,ijc (4) where: MTS = Math test scores RTS = Reading test scores ISO = Indicator of Sociability ISE = Indicator of Self-esteem ICD = Indicators of classroom democracy X, W, G, Z = vector of control (exogenous) variables α,β,γ,δ,θ,π, ,ω are structural parameters to be estimated ε1,…,ε4 are disturbance or error terms i, j, c represent the ith pupil, the jth school, and cth country respectively The explanatory endogenous variables are correlated with the error terms in the system of equations. All oth- er explanatory variables are considered exogenous to the system and uncorrelated with the error terms. The exogenous variables are considered as instruments for the endogenous variables. Njong, Classroom Democracy and Learning Outcomes 51 Model Estimation To estimate a system of simultaneous equations, instrumental variable methods for joint estimation are often proposed. To account for the nested structure of the data due to intra-school correlations that may arise because responses from children in the same school are likely to be similar while those from children in other schools are different, the solution in single equation model is to cluster the errors at the level of the school (Fielding, 2010; Wamalwa & Burns, 2018). To achieve this within a simultaneous equation framework, Baltagi & Chang (2000) and Hsiao (2003) proposed an error components-two stage least squares (ec2sls) estimator to account for the random error component of the sys- tem of equations. The ec2sls estimator is implemented on an equation by equation estimation basis (Baltagi, 2008)5 to avoid having inconsistent equation estimates for all the equations in the event where any of the equa- tions is mis-specified. Pair-wise Correlation Results Table 1 shows the weighted6 pair-wise correlations between the classroom democratic indicators and learning outcomes as defined in this study. Table 1 shows some preliminary partial inter-re- lationships among learning outcomes and classroom democratic indicators. First observation on Table 1 is that there is a high positive correlation between the cognitive educational outputs. The correlation coef- ficient between math skills and reading test scores is 0.845 indicating that pupils who read well report high performance in mathematics as well. Second interesting observation is that correlation between non-cognitive outcomes is moderate and positive. For instance, the correlation coefficient between self-esteem and socia- bility is 0.59, indicating that that children who have high self-esteem are much likely to be sociable. Third observation of interest is that the correlation between the cognitive and the non-cognitive indicators are pos- itive but much lower. For example, correlation coeffi- cient between self-esteem and reading is 0.201, while it is 0.155 with math. The correlations among the indica- tors of classroom democracy are very low. Estimation Results In this section the regression results of classroom democratic practices on learning outcomes are pre- sented. For convenience, the results for cognitive skills are presented first. Table 2 (see next page) displays the estimation results for cognitive skills. The results on Table 2 show high positive and sig- nificant correlations between both test scores at the 1% level. This result suggests that a child who does well in reading generally will also achieve well in math, and vice versa. The other two endogenous variables; self-es- teem and sociability also show significant positive, but much lower correlations with math and reading skills. As concerns the indicators of classroom democracy, the correlations are positive and insignificant, except for child discipline. The results are interpreted in terms of standard deviations (SDs). The indicator of group work correlates positively with cognitive skills. This means that a one unit increase in the indicator of group work improves math and reading scores by .073 SDs and .065 SDs respectively, but the effects are insignif- icant. The results indicate that when children’s rights increase by one unit, math and reading test scores rise insignificantly by .068 and .052 SDs respectively. The indicator of child discipline associates positively and significantly with cognitive skills (.052 SDs for math and .083 SDs for reading). The indicator of cultural identity indicates that when it increases by one unit, math scores improve by .061 SDs; while reading scores increase by .071 SDs. Of the control variables, the following are positive and significant at least at the 5% level: household size, teacher experience, teacher’s training, community in- volvement in school management, classroom pedagogic resources, and school infrastructure. The models pre- dict that higher test scores are associated with students who are male, though this finding is insignificant. Global Journal of Transformative Education (2020) Vol 2 5Baltagi's (2008) ec2sls can be obtained by specifying the ec2sls option, after xtivreg in Stata. 6The PASEC survey provides several types of sampling weights, notably child, teacher, or school weights depending on the level of analysis. Since unit of analysis in this study is the child, I use child sampling weights. Njong, Classroom Democracy and Learning Outcomes 52 Teacher absenteeism has negative significant effects on academic achievements. The diagnostics statistics on Table 2 show that the vari- ables specified in the models accounted for approximately 27.4% of the total variance of the math scores, and 37.1% of reading test scores, respectively. The between variations (R-sq=56.8% for math; 59.3% for reading) accounted more than the within variations (R-sq=22.1% for math; 32.7% for reading) in explaining the overall variances of the models, respectively. The F-values are statis- tically significant at the 1% level. This indicates that the regression models fit the data structure. In other words, the estimates of each model taken together are significant. Table 3 displays the estimation results for non-cognitive skills. The results on Table 3 show a moderately positive and significant association between self-esteem and sociability. Specifically, a unit increase in self-esteem improves sociability by .051 SDs. This suggests that a child’s feeling of self-esteem is affected by their inter- est in other children and the reactions they receive in return. The other two endogenous variables; math and reading show much lower positive and significant cor- relations with self-esteem and sociability skills. As regards the indicators of classroom democracy, the associations are positive and insignificant, except for child discipline. The indicator of group work is shown to associate positively and insignificantly with self-esteem (0.011 SDs) and sociability (.057 SDs). As concerns the indicator of children’s rights, Table 3 shows that a one unit increase in children’s rights improves self-esteem and sociability insignificantly by .013 and .035 SDs respectively. The indicator of child discipline associates positively and significantly with non-cognitive skills (.022 SDs for self-esteem and .024 SDs for sociability). Table 3 also indicates that when indicator of cultural identity increases by one unit, self-esteem improves by .026 SDs; while sociability ameliorates by .006 SDs, though these improvements are insignificant. As concerns the other exogenous correlates, the following relate positively and significantly with the non-cognitive outcomes: household size, socio-eco- nomic status, teacher experience-squared, community involvement in school management, and classroom pedagogic resources. The models predict school infra- structure, teacher’s professional training, and experi- ence associate positively though insignificantly with non-cognitive skill. As expected, teacher absenteeism has negative, though insignificant effect on non-cogni- tive achievements. Precisely, a unit increase in teacher absenteeism causes self-esteem to change by -.004 SDs, while sociability changes by -.022 SDs. The diagnostics tests on Table 3 show that the variables specified in the models accounted for approx- imately 29.1% of the total variance for self-esteem, and Global Journal of Transformative Education (2020) Vol 2 Njong, Classroom Democracy and Learning Outcomes 53 Discussion The estimation results of this study show positive and significant inter-relationships between the learning outcomes of math, reading, self-esteem and sociability. In an attempt to explain these positive correlations, I draw from the theory of self-determination (Deci, Vallerand, Pelletier & Ryan, 1991). According to this theory intrinsically motivated individuals will engage in actions or behaviours because they enjoy doing them. In this case the individual has an internal motivation and feels fully engaged to undertake the action. This contrasts with externally motivated behaviours that individuals may initiate because they expect a reward or want to avoid a punishment. Thus, it is expected that pupils who are intrinsically motivated (have high self-esteem and sociability) exhibit higher conceptual understanding (Deci, Vallerand, Pelletier & Ryan, 1991). Such children are more confident in their capacities and therefore have greater interest for learning. These results corroborate the study by Noftle & Robins (2007) who found positive association between the person- ality dimensions of students and their math GPA and verbal test scores. Chamorro-Premuzic & Furnham (2003) have also reported positive significant correla- tion between personality traits and student academic performance. The estimation results of this study portray positive correlations between all the indicators of classroom democracy and learning outcomes in the context of Francophone African countries. However, only the association between child discipline and the learning outcomes is significant, the correlations of the other indicators of classroom democracy with educational achievements are insignificant. The positive associa- tion between indicator of group work and educational outcomes means that when pupils are given the oppor- tunity to work in small groups, they are more likely to have higher test scores. The positive association be- tween indicator of group work and learning outcomes corroborates findings from some earlier studies such as Feinberg & Toress (2001) and Lipton & Oakes (2003) who found that when pupils participate actively in class and work as teams, their educational achievements are much higher. Unfortunately, our positive correlation is insignificant. In the survey about 63.2% of the teachers reported not organising pupils to work in groups. Fur- thermore, about 67.3% of the teachers lacked the tech- niques to do so (see Table S1 in the Appendix). Children in the African family setting share responsibilities in carrying out household chores such as cooking, taking 29.9% of sociability, respectively. The between variations (R-sq=47.3% for self-esteem; 47.5% for sociability) accounted more than the within variations (R-sq=26.7% for self-esteem; 27.0% for sociability) in explaining the overall variances of the models, respec- tively. The F-statistics indicate that the regression mod- els fit the data structure. Global Journal of Transformative Education (2020) Vol 2 Njong, Classroom Democracy and Learning Outcomes 54 care of babies, fetching firewood and water, and run- ning errands for the elderly. Thus at an early age Afri- can children are exposed to a ‘team-spirit way’ of living which unfortunately is disrupted in the classroom when the teacher fails to organise them to work in groups. Children will learn better in a classroom atmosphere, where children work in smaller groups, feel safe to express their feelings and open discussions are encour- aged. Therefore, there is need in African countries to move away from the traditional mode of teaching and learning where children listen and teachers speak and dictate notes. The estimation results indicate positive but insignif- icant correlations between indicator of children’s rights and learning outcomes. As a follow up of the 1989 UN Convention on children’s right, a special session was held on 22 May 2002 on children during which coun- tries committed to invest in children and adopt policies in the best interest of the child. Francophone African countries have several developmental challenges such as low incomes, inadequate access to portable water and sanitation, food insecurity and hunger, and wide- spread disease that threaten children’s livelihoods (Watkins, 2016). Given these challenges it is no surprise that 57.5% of the pupils in the survey reported feeling hungry in class, more than 77- 83% reported having one health impairment or another and 73.6% did not attain nursery school. A further 36% of the children have never slept under a mosquito net (see Table S2). These are basic rights of the child. The inability of the children’s families to respect these rights may account for the insignificant nature of the results. It would be a miracle to expect children to achieve significantly when such basic rights are not respected. In tandem with African culture children in poorer families may be re- quired to carry out petty commercial activities, or join their parents in farming so as to help the family survive. In western culture this may be considered child labour or child abuse, but in the African context this is not only a livelihood survival strategy but is often considered a normal activity in line with the household chores distri- bution plan of the family. Thus, I may further argue that differences of this nature in the conceptualisation of children’s rights may also account for the insignificance of the estimation result. The indicator of child discipline associates positive- ly and significantly with learning outcomes. This result is in line with expectations and agrees with Bodovski, Nahum-Shani, & Walsh (2013), which showed that higher math scores were registered among US students in a disciplinary school environment. Further, Ning, Van-Damme, Yang, & Gielen, (2013) also obtain a posi- tive and significant correlation between child discipline and reading achievements. Furthermore, Pasternak (2013) and Zhao & Kuo (2015) have reported posi- tive and significant relationships between children’s self-discipline and non-cognitive achievements. Some unique features of the African traditional society may account for the significance of the child discipline indicator in this study. The extended family system is a powerful tool in parenting and helps to develop a sense of social responsibility, cooperation, and respect for elders in the child’s early years (Nsamenang, 2008). Another feature in the African traditional society aspect is the telling of folktales. This is another form of par- enting which inculcates good morals and discipline in children (Degbey, 2012). Given these indigenous cultural values, the African child at an early age is more likely to be disciplined and by extension more likely to respect classroom rules and regulations. The indicator of cultural identity indicates a pos- itive but insignificant association with learning out- comes. This finding slightly differs from results of earlier studies, like Okonji (1971); Bodrova & Leong (2007); Marfo & Biersteker (2011) which find that cultural values have positive and significant effects on children’s academic performance. A close look at the items in the 2014 PASEC survey from which the indi- cator of cultural identity is constructed may provide some explanation for the insignificance of estimation result. Only about 18.9% of the teachers expressed themselves in the native language of the pupils where the school is located; 70.7% of the teachers did not allow pupils to speak, and sing in class in their native language and worst still only 8.6% of the pupils speak exclusively the mother tongue at home (see Table S4). These attitudes are not of a nature to facilitate learning and ameliorate academic achievements in the African context. Mucherah & Mbogori (2019) argue that Afri- can children will learn better if the native language and French/English are used simultaneously as languages of instructions in school. This will facilitate the use of local examples to introduce the alphabet and counting of objects. Unfortunately, Francophone African scholars are still caught in the empty pride of using examples from Paris, France, during lessons in class. African children may not quickly figure out such examples in their local context. Some parents are also still entangled in the web of colonisation, and feel more westernised if their children speak only French/English. African Global Journal of Transformative Education (2020) Vol 2 Njong, Classroom Democracy and Learning Outcomes 55 Global Journal of Transformative Education (2020) Vol 2 educationists have to encourage a shift away from these colonial mental enslavement and practices. Conclusion and Policy Implications Conclusion This study attempts to shed light on the effects of classroom democratic practices on cognitive (math and reading) and non-cognitive (self-esteem and sociabili- ty) learning outcomes using cross-sectional data from Francophone African countries. To capture the in- ter-relationships inherent in the education production process, a simultaneous equation modeling approach is proposed. In order to estimate the model parameters, an error component- two stage least squares-instru- mental variable method is used to properly account for the multilevel nature of the data structure. The results confirm the endogenous nature of the learning out- come variables; specifically, they indicate positive and significant correlations among the learning outcomes. As regards classroom democratic practices, the results indicate positive and insignificant associations with the dimensions of group work, children’s rights, and cultural identity. Only the indicator of child discipline has positive and significant effect on learning outcome skills in this study. The exogenous or control variables included in the model have significant expected signs, notably family size, socio-economic status, teacher experience, community involvement, and classroom pedagogic resources. Policy Implications From the findings of this study, it stands out that teachers and parents are some of the most important stakeholders in the process of having children inter- nalize democracy in the classroom. In this regard, the following policy recommendations are made: Teachers should initiate and encourage children to work in small groups in the classroom. Teachers should develop the habit to invite pupils to always see them- selves as capable and valuable. This action will bring out the positive traits and potentials that the children have and help to build their self-esteem which is a strong internal motivation to educational achievement. Education stakeholders should take measures to pre- serve and promote indigenous cultural values because they tend to facilitate learning achievements. Education policy makers should devote more resources to revise the curriculum of professional colleges in order to train teachers to: i) develop skills of assigning and supervi- sion of pupils in team work; ii) be culturally responsive in their teaching practices and accommodate pupils who speak their native language, or display their cul- tural identity in class. Therefore, the teacher would be acting like a classroom democratic model. Education- ists should develop parenting programmes to sensitise parents to speak their native languages with children at home, and how to better prepare children for schooling as regards nutrition, health, and other nurturing issues. The findings also showed that there are variations in achievements within schools and between schools. The implication of this result is that educationists must design multiple teaching strategies. A blind spot in the education system is an attempt to treat all children the same. This certainly is for reasons of equity and fair- ness, but in reality, children are all different given that they come from diverse backgrounds. Njong, Classroom Democracy and Learning Outcomes 56 References Anderson, L.W., Krathwohl, D.R., Airasian, P.W., Cruik- shank, K.A., Mayer, R.E., Pintrich, P.R., Raths, J., & Wittrock, M.C. (2001). A taxonomy for learning, teaching, and assessing: A revision of Bloom’s Taxon- omy of Educational Objectives. New York: Pearson, Allyn and Bacon. Baltagi, B. H. (2008). Econometric analysis of panel data. (4th Ed.). New York: Wiley. Baltagi, B. H. (2011). Simultaneous equations model. (5th Ed.). Berlin, Heidelberg: Springer. Baltagi, B. H., & Chang, Y. J. (2000). Simultaneous equa- tions with incomplete panels. Econometric Theory, 16(2), 269-279. Bandura, A. (1977). Social learning Theory. New York: General Learning Press. Barron, B., & Darling-Hammond, L. (2008). Teaching for meaningful learning: A review of research on inqui- ry-based and cooperative learning. Stanford, CA: Stanford University. Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K. D. (2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Pub- lic Interest, 4(1), 1-44. Bodovski, K., Nahum-Shani, I. & Walsh, R. (2013). School disciplinary climate and students’ early mathematics learning: Another search for contex- tual effects? American Journal of Education, 119(2), 209-234. Bodrova, E., & Leong, D.J. (2007). Tools of the mind: The Vygotskian approach to early childhood education. (2nd Ed). Upper Saddle River, NJ: Merrill/Prentice Hall. Borghans, L., Duckworth, A. L., Heckman, J. J., & Ter Weel, B. (2008). The economics and psychology of personality traits. Journal of Human Resources, 43(4), 972-1059. Chamorro-Premuzic, T., & Furnham, A. (2003). Per- sonality predicts academic performance: Evidence from two longitudinal university samples. Journal of Research in Personality, 37(4), 319-338. Deci, E. L., Vallerand, R. J., Pelletier, L. G., & Ryan, R. M. (1991). Motivation and education: The self-deter- mination perspective. Educational Psychologist, 26, 325-346. Degbey, J. L. (2012). Africa family structure. Retrieved from http://www.jicef.or.jp/wahec/ful217.htm Emler, N. (2001). Self esteem: The costs and causes of low self worth. York Publishing Services. Fearnley-Sander, M., Moss, J., & Harbon, L. (2001). The civic school: Australian-Indonesian professional collaboration to model and audit the development of democratic primary classrooms and teacher lan- guage using the Index for Inclusion. Paper presented at the Australian Association Research in Education (AARE) Conference, Sydney. Feinberg, W., & Torres, C. A. (2001). Democracy and education: John Dewey and Paulo Freire. Education and Society, 59-70. Fielding, A. (2010). Module 8: Multilevel modelling in practice: Research questions, data preparation and analysis. LEMMA VLE, University of Bristol, Centre for Multilevel Modelling. Glewwe, P. (2002). Schools and skills in developing countries: Education policies and socioeconomic outcomes. Journal of Economic Literature, 40(2), 436-482. Glewwe, P. & Jacoby, H. (1994). Student achievement and schooling choice in low-income countries: Evidence from Ghana. Journal of Human Resources, 843-864. Glewwe, P., & Kremer, M. (2006). Schools, teachers, and education outcomes in developing countries. Hand- book of the Economics of Education, 2, 945-1017. Greene, W. H. (2011). Econometric analysis. (7th Ed.) Upper Saddle River, NJ: Prentice Hall. Harter, S. (2012). The construction of the self. (2nd ed.). New York: Wiley. Hsiao, C. (2003). Analysis of Panel Data (2 bs.). İngiltere: Cambridge University. Levin, J., & Nolan, J. F. (1991). Principles of classroom management: A hierarchical approach. Prentice Hall. Lipton, M., & Oakes, J. (2003). Teaching to change the world (2nd ed.). Boston: McGraw Hill. Maftoon, P., & Shakouri, N. (2012). The concept of power in teacher talk: A critical discourse analysis. World Applied Sciences Journal, 19(8), 1208-1215. doi: 10.5829/idosi.wasj.2012.19.08.1894 Marfo, K., & Biersteker, L. (2011). Exploring culture, play and early childhood education practice in Afri- can contexts. In S. Rogers (Ed.), Rethinking play and pedagogy in early childhood education (pp. 73–85). New York, NY: Routledge. Global Journal of Transformative Education (2020) Vol 2 http://eric.ed.gov/?id=EJ1003815 http://eric.ed.gov/?id=EJ1003815 http://eric.ed.gov/?id=EJ1003815 Njong, Classroom Democracy and Learning Outcomes 57 Global Journal of Transformative Education (2020) Vol 2 Marfo, K., Pence, A. R., LeVine, R. A., & LeVine, S. (2011). Strengthening Africa’s contributions to child devel- opment research: Introduction. Child Development Perspectives, 5(2), 104–111. Mothata, W. & Squelch, J. (1997). Eight keys to effective school management in South Africa. Halfway House: Southern Book Publishers. Mucherah, W. & Mbogori, T. (2019). Examining child de- velopment from an African cultural context. Global Journal of Transformative Education, 1(1), 11-17. doi: 10.14434/gjte.v1i1.26140 Nagaraju, C., Madhavaiah, G., & Peter, S. (2013). Teach- er-centred learning and student-centred learning in English classroom: The teaching methods realiz- ing the dreams of language learners. International Journal of Scientific Research and Reviews, 2(3), 125-131. Ning, B., Van -Damme, J., Yang , X., & Gielen, S. (2013). September). Does classroom disciplinary climate in a school matter everywhere? A cross-country com- parative study. Paper presented at The 2013 Annual Meeting of the European Conference on Educational Research, Istanbul, Turkey. Njong, A. M., & Ningaye, P. (2008). Characterizing weights in the measurement of multidimensional poverty: An application of data-driven approaches to Cameroonian data. OPHI Working Paper No. 21. Noftle, E. E., & Robins, R. W. (2007). Personality pre- dictors of academic outcomes: Big five correlates of GPA and SAT scores. Journal of Personality and Social Psychology, 93(1), 116. Nsamenang, A.B. (2008). Constructing cultural identity in families, In: Brooker, L. and Woodhead, M. (Eds.). Social inclusion and respect for diversity: Develop- ing positive identities. Milton Keynes, Buckingham- shire, UK: The Open University. Okonji, M. O. (1971). The effects of familiarity on clas- sification. Journal of Cross-Cultural Psychology, 2, 39–49. Olsen, J. M., Breckler, S. J., & Wiggens, E. C. (2008). Social psychology alive. Toronto: Thomson Nelson. PASEC (1998). L’enseignement primaire au Cameroun: investigations et diagnostics pour l’amélioration de la qualité du système éducatif. PASEC, CONFEMEN, Dakar. PASEC (2017). Manuel d’exploitation des données: Éval- uation internationale PASEC2014. PASEC, CONFE- MEN, Dakar. Papadopoulos, K., Metsiou, K., & Agaliotis, I. (2011). Adaptive behaivor of children and adolescents with visual impairments. Research in Developmental Dis- abilities, 32, 1086–1096. Pasternak, R. (2013). Discipline, learning skills and ac- ademic achievement. Journal of Arts and Education, 1(1), 1-11. Saigal, S., Lambert, M., Russ, C., & Hoult, L. (2002). Self-esteem of adolescents who were born prema- turely. Pediatrics, 109, 429–433. Serpell, R., & Marfo, K. (2014). Some growth points in African child development research. in R. Serpell & K. Marfo (Eds.), Child development in Africa: Views from inside. New Directions for Child and Adolescent Development, 146, 97–112. UN General Assembly. (1989). Convention on the Rights of the Child. United Nations, Treaty Series, 1577(3). Wamalwa, F. M., & Burns, J. (2018). Teacher Human Capital, Teacher Effort and Student Achievements in Kenya (No. 722). Watkins, K. (2016). The State of the World’s Children 2016: A Fair Chance for Every Child. New York: UNICEF. Woolridge, J. M. (2009). Introductory cconometrics: A modern approach. (4. Ed). South-Western, Michigan State University, 378, 57. Zhao, R., & Kuo, Y. L. (2015). The role of self-discipline in predicting achievement for 10th graders. In- ternational Journal of Intelligent Technologies and Applied Statistics, 8(1), 61-70. doi: 10.6148/IJI- TAS.2015.0801.05 Aloysius Mom Njong (mom_aloys@yahoo.fr) is a Professor of Development Economics at the University of Bamenda, Cameroon. Author https://scholarworks.iu.edu/journals/index.php/gjte/article/download/26140/32308/ https://scholarworks.iu.edu/journals/index.php/gjte/article/download/26140/32308/ https://lirias.kuleuven.be/handle/123456789/428773 https://lirias.kuleuven.be/handle/123456789/428773 https://lirias.kuleuven.be/handle/123456789/428773 https://lirias.kuleuven.be/handle/123456789/428773 https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://ora.ox.ac.uk/objects/uuid:8d71f02f-9698-4453-8ffb-0b40361ac2fd/download_file?safe_filename=Characterizing%2Bweights%2Bin%2Bthe%2Bmeasurement%2Bof%2Bmultidimensional%2Bpoverty%2B21.pdf&file_format=application%2Fpdf&type_of_work=Working+paper https://www.careperinatologia.it/lavori/L52.pdf https://www.careperinatologia.it/lavori/L52.pdf https://www.econrsa.org/system/files/publications/working_papers/working_paper_723.pdf https://www.econrsa.org/system/files/publications/working_papers/working_paper_723.pdf https://www.econrsa.org/system/files/publications/working_papers/working_paper_723.pdf Njong, Classroom Democracy and Learning Outcomes 58 Global Journal of Transformative Education (2020) Vol 2 Table S1n Table S2n Table S3n Njong, Classroom Democracy and Learning Outcomes 59 Global Journal of Transformative Education (2020) Vol 2 Table S4n Table S5n Table S6n