Science Epistemological Beliefs from the Perspective of Culture and Gender in Co-Educational Secondary Schools, Kenya

The purpose of the study was to investigate the effects of culture and gender on the development of science epistemological beliefs among co-educational secondary schools in Kisii and Homa-bay counties in Kenya. The study employed a causal comparative design with purposive sampling technique. A 26-item questionnaire called Science Epistemological Beliefs Questionnaire (SEBQ) developed by Conley et al. (2004) was used. The instrument assesses four dimensions of Epistemological Beliefs as Source, Certainty, Development and Justification of scientific knowledge. It was administered to 410 students (Kisii-Abagusii, n = 207; Homabay-Luo, n= 203). The data were analyzed by culture and gender using independent sample t-tests. The findings indicate statistically significant cultural differences in favor of the Abagusii students for three dimensions (Source, Certainty and Development) and statistically significant difference in favor of the Luo students for the dimension of Justification. The findings also indicate non-significant gender differences for the dimensions Source, Certainty and Justification. On the other hand there were statistically significant gender differences in favor of the boys for the dimension of development. Discussions, conclusions and implications are herein made.


Introduction
In a century where great premium is placed on knowledge transfer to novel situations, many educational researchers have given much attention to students' Mohamed & El-Habbal, 2013). According to Hofer and Pintrich (1997), epistemological beliefs are beliefs about the nature of knowledge, knowledge acquisition and knowing. Schommer (1990) defined epistemological beliefs as individual's subjective beliefs of the definition of knowledge and the way in which the process of acquiring knowledge takes place. Implicit in these definitions is the fact that epistemological beliefs determine how knowledge is acquired by learners and whether that knowledge will be transferable to novel situations.
According to Lodewyk (2007) the early theories of personal epistemology characterized the concept as relatively one-dimensional having a rather uniform developmental trajectory. Perry (1970) argued that students who enter a college perceive knowledge to be simple, certain and provided by an instructor. However as they leave, the same students hold more sophisticated beliefs, viewing knowledge as complex, tentative and derived from a logical and empirical process. Kuhn, Cheney, & Weinstock (2000) distinguished three developmental stages of epistemological beliefs as absolutism (recognition of knowledge as certain and absolute), multiplism (cognizance of knowledge as uncertain and subjective) and evaluatism (acknowledgement of both uncertainty of knowledge and significance of evaluating knowledge claims). Chandler et al. (cited in Hofer, 2008) recognized an additional stage earlier than absolutism referred to as naïve realism among very young children in which their own perception of knowing is the only view accessible to them. Arslantas (2016), citing the findings of a 5-year longitudinal study by Magolda (1992) on epistemological beliefs explicates four dimensions of epistemological beliefs as absolute, transitional, independent and commitment. Ricco, Pierce, & Medinilla (2010) observe that the stage based model of personal epistemology indicates that a movement from an objectivist and absolutist epistemology to a more subjectivist and relativist stance of epistemology can occur during pre-adolescence and even middle childhood. In this theoretical positioning, Chan and Elliot (2004) have opined that an individuals' belief system comprises clusters of beliefs interacting with one another and influencing one's decision making, attitudes and actions. Lodewyk (2007) holds that a general theme in epistemological theory is that students develop from a position where knowledge is perceived as right or wrong to a position of relativism and then to a position in which individuals are active constructors of meaning. This perspective recognizes epistemological beliefs as existing in a continuum. However, recent theories have emphasized the multidimensionality and variable developmental path of the construct (Lodewyk, 2007). This implies that epistemological beliefs constitute a sub-system within one's belief system and are considered to have multiple dimensions. Schommer (1990) explained epistemo-logical beliefs with a model of five dimensions which included Stability (tentative to unchanging), Structure (isolated to integrity), Source (authority to observation and reason), Speed of acquisition (quick or gradual), Control of acquisition (fixed at birth or lifelong improvement). Many other models have arisen from Schommer's dimensions. This study proceeds in the philosophical trajectory that these beliefs have multiple dimensions and develop in asynchronous fashion.
Much attention has been directed at studying the relationships between epistemological beliefs and self-regulated learning strategies (Savoji, Niusha, & Boreiri, 2013), Achievement motivation (Ricco, Pierce, & Medinilla, 2010), Academic achievement (Mohamed & El-Habbal, 2013;Barvarz, Nami, & Ahmadi, 2014;Arslantas, 2016). On the other hand, these beliefs have implications for learning, cognitive processing, conceptual change, strategy use and pedagogy (Hofer & Pintrich, 2002;Hofer, 2008;Chan & Elliot, 2004;Arslantas, 2016;Berding, Rolf-Wittlake, & Buschenlange, 2017). Hofer (2008) observed that research across cultures on epistemological beliefs could help to advance an understanding of the epistemological development trajectory across cultures. These conceptions of epistemological beliefs imply that they are evolutionary and dynamic in nature, implying that individuals' belief about the nature of knowledge and how it is constructed keeps changing. This consequently implies that epistemological beliefs held by individuals affect how they learn and the strategies which they adopt to learn. Culture could also be the lens through which an understanding of epistemological beliefs is expanded. Chen and Pajares (2010) have opined that although much research has been done examining gender differences in epistemological beliefs, some have found significant differences. Chen and Pajares (2010) continue to aver that gender may play an important role in epistemological reasoning in ways that are undetectable. Karabenick and Moosa (2005) investigated science epistemological beliefs from the perspective of gender and culture among Omani and US college students. The findings indicated that the Omani more than US students believed that knowledge in the sciences was simpler, more certain and accepted scientific authorities as the basis of scientific truth. On the other hand, Omani men were more accepting of authorities than women but the US students exhibited no gender differences. A study by Conley et al. (2004) revealed that students became more sophisticated in their beliefs about source and certainty of knowledge over time but there were no reliable changes in development and justification. There were also no ethnic and gender differences in epistemological beliefs. In a study Lodewyk (2007) examined students' epistemological beliefs according to gender, school orientation, overall academic achievement and performance on two differently structured academic tasks. The findings indicated that belief in simple knowledge significantly predicted overall performance and reflective judgment scores on the ill structured tasks but not on the well-structured tasks. Peer (2005) investigated Singaporean students' epistemological beliefs from the perspective of gender using epistemological beliefs questionnaire by Conley et al. (2004). The findings indicated that the mean scores of the female students in all the four subscales were slightly higher than boys' though the differences were not statistically significant. Cano (2005) analyzed changes in epistemological beliefs of boys and girls in secondary schools as they progressed through their studies using Schommer's epistemological beliefs questionnaire. The findings indicated that the boys' beliefs in "quick learning" are more unstable throughout secondary education than girls. The beliefs in "simple knowledge" and "certain knowledge" however followed a similar pattern of development. Marzooghi, Fouladehang and Shemshiri (2008) carried out a study to investigate gender differences in epistemological beliefs using epistemological beliefs questionnaire. The results showed males had naïve epistemological beliefs than females. The girls were higher than boys in terms of their thinking about nature of learning, source of knowledge or the structure of knowledge. Yenice (2015) investigated epistemological beliefs of Turkey students using Schommer's scale which was adapted into Turkish language. The findings indicated that gender does not have a significant effect on the factor "Belief in Learning based on attempt" and the factor "Belief of there is one truth". However, the variable gender was found to have a significant effect on the factor "Belief on learning based on ability". Arslantas (2016) found out no gender differences on teachers candidates' epistemological beliefs on any of the scales of the measure. There were statistically significant relationship between teacher candidates' grade point average and the sub-scale of learning being dependent on talent. Zhu, Valcke, & Schellens (2008) carried out a study to investigate cultural differences in epistemological beliefs using Schommer's epistemological beliefs questionnaire. The findings indicated cultural differences between Chinese and Flemish students in terms of "source of knowledge" and "ability to learn" on the other hand, there were similarities in terms of belief of "certain knowledge" and "fixed ability". Filoteo, Talisayon, & Ferido (2014) carried out a study to find out cultural/ethnic differences between Chavacanos, Tausugs and Visayans from an experimental perspective. Before the enhanced epistemological instructional intervention, there were slight differences in epistemological beliefs however not statistically significant. After the intervention, there were no differences in their epistemological beliefs in Chemistry. Chen and Pajares (2010) investigated ethnic differences in science epistemological among Asian, white students, African

Literature Review
American and Hispanic students using the Science epistemological beliefs questionnaire by Conley et al. (2004). The findings indicated that when controlling for previous achievement, the Hispanic students ascribed to more naive beliefs about "certainty" of scientific knowledge compared to their Asian, White and African American peers. Hispanic students and Asian students ascribed to more naive beliefs about "source" of scientific knowledge compared to their African American and White peers. Epistemological beliefs about the "Development" or "Justification" of knowledge were not statistically significant. Tabak and Weins-  Lin et al. (2013) explored the differences in high school students' science epistemological beliefs and the different relationships between them in Mainland China and Tai-wan. The findings indicated that the Taiwanese students generally were more prone to believe that scientific reality is invented, that the development of scientific knowledge is always changing and its status is tentative than the Chinese students were.
The literature reviewed above indicates that studies on epistemological beliefs in relation to culture/ethnicity needs extension to more cultures since differences exist in almost every comparison. The findings on gender based studies are mixed, inconclusive and unequivocal. This implies continuous studies need to be carried out to get a more comprehensive picture of epistemological beliefs from the perspective of gender.

Cultural Contexts of the Study
According to Watson, Ho, & Raman (1994) culture is the beliefs, values, systems, norms, mores, myths and structural elements of a given organization, tribe or society. They continue to observe that culture exists at national, regional and at corporate levels. Klassen (2004) has observed that a more individualistic culture may be more optimistic in predicting their ability or performance whereas people from a more collectivist cultures may be more realistic in predicting their ability or performance. In the Abagusii culture, men are allowed to be polygamous and exercise control and authority over the rest of the family members. The men have exclusive rights over the control of land, property. The norms and the customs of the Abagusii advantage men in the society and at the same time it encourages social cohesion (Ndeda, 2019). The Luo culture on the other hand, has closely linked economic and social systems that continue to evolve as they migrate.
The Luo also practiced polygamy that was a desirable index of a man's worth and an indication of a man's wealth and enhanced his economic and political status. For a woman, life was characterized by competition with co-wives (Ndeda, 2019). Ndeda continues to opine that women control their households where they plant the concept collective ownership and antagonism against the other households. In this scenario, wives become rivals in the possession of property and competition for household favor which promotes self-pride in the women.
Division of labour was based on gender and age, however despite this, there was evidence of equality between the genders with regard to their daily work. Using the Hofstede lens to analyze these cultural orientations, it appears that the Abagusii culture enhances a high power distance whereas the Luo culture enhanced High individualism (Hofstede, 1980

The Present Study
Studies on epistemological beliefs have dominated the many parts of the world due to its significance in the process of learning and because there is a relation-

Research Questions
The study was guided by the following questions:-1) Are there cultural differences in Science Epistemological Beliefs of Abagusii students and Luo students?
2) Are there Gender differences in Science Epistemological Beliefs among the students in the co-educational secondary schools?

Methodology
This section presents the research design that was adopted for the study, the sample of the study, instrumentation and methods of data analysis.

Research Design
The study adopted a causal-comparative design. Causal-comparative research does not involve control of variables because their effects have occurred or they have fixed manipulative and uninfluenceable properties (Mugenda & Mugenda, 2003;Gall, Borg, & Gall, 2003). Causal comparative research attempts to enquire the cause and effect connection for pre-existing differences in groups of individuals (Fraenkel & Wallen, 2008). The researcher therefore examined the effects of culture and gender (independent variables) on the development of science epistemological beliefs (dependent variables). The consequences in the groups had already occurred hence the independent variables were not manipulated. The groups were compared in order to investigate if there were significant differences in science epistemological beliefs. The possible reasons for the differences in culture and gender were also given (Gall, Borg, & Gall, 2003).

Sampling Procedures and Sample Size
Two schools that had the needed attributes were purposively sampled for the study. Purposive sampling technique is suitable where attributes of the population are familiar or information rich with respect to the purpose of the study (Mugenda & Mugenda, 2003;Gall, Borg, & Gall, 2003;Fraenkel & Wallen, 2008). Kisii County is dominated by the Abagusii community and Homabay County is dominated by the Luo community, hence had the required respondents for the study. In this study, two pre-existing samples of students from Kisii County and Homabay County co-educational schools were selected in the pursuit of investigating and determining their science epistemological beliefs. Sampling was cautiously done to ensure that the schools with the characteristics of culture and gender parity in enrolment are included in the study. One school was sampled from Homabay County and another one from Kisii County. The sample for this study was 410 students. Table 1 gives the details of sample characteristics.

Instrumentation
The science epistemological beliefs were measured using Science Epistemological Beliefs Questionnaire (SEBQ). This instrument was developed by Conley et al. (2004) to measure science epistemological beliefs. The instrument is a 26-Item questionnaire with 4 subscales as source (5 items), certainty (6 items), development (6 items) and justification (9 items). The items were rated on a 5-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree). The instrument had an overall reliability of 0.73. A level of reliability of 0.7 and above is considered appropriate and acceptable for studies (Ogunniyi, 1992;Gall, Borg, & Gall, 2003). The items for the source and certainty scales were reversed, consequently, scoring was reversed to reflect this. The instrument was administered to students in the two schools and in all the grades by the researcher assisted by the science teachers in the sampled schools. The administration of the instrument took 30 minutes. The data collection process took place in second term of the Kenyan school calendar. The scales and their descriptions are shown in Table 2.

Data Analysis
The data from SEBQ were analyzed according to the sub-scales of the instrument. Each item in the instrument was worth a minimum of one point and a maximum of 5 points. Hence the minimum mark that a student could score was 26 and a maximum of 130 for all the sub-scales. Descriptive statistics were used to summarize raw data according to the objectives of the study. Inferential statistics were used to analyze quantitative data and test the research hypotheses.
The hypotheses were accepted at a significance level of α = 0.05. To determine the effect of culture and gender on the development of science epistemological beliefs, independent sample t-tests were carried out. The t-test is an inferential statistical procedure used to determine whether means of two samples are significantly different (Fraenkel & Wallen, 2008). The dependent variable (science epistemological beliefs) was data in ratio scale, the groups were mutually exclusive, there were relationships between observations in each group and the independent variables (ethnicity and gender) were two categorical independent groups. In this regard, independent sample t-test was appropriate for this analysis (Gall, Borg, & Gall, 2003). Data analysis was conducted with the aid of Statistical Package for Social Sciences (SPSS) version 23.

Results
The results of the study are presented according to the objectives by first of all presenting descriptive statistics followed by inferential statistics.

Cultural Differences in Science Epistemological Beliefs
To determine the effect of culture on the development of science epistemological beliefs, the descriptive data analysis are first presented followed by t-test data analysis. Table 3 shows the descriptive statistics. Table 3 indicates that all students had scores above the mid-point of 12.5 for source, 15 for certainty and development and 22.5 for justification. On the other hand, Abagusii students had higher mean scores in all the sub-scales (Source, Certainty and Development) except in the scale of Justification where the Luo students had higher scores. To determine whether these differences were statistically significant, independent sample t-tests were carried out. Table 4 shows the output of Levene's test and independent sample t-test for the various subscales of SEBQ.
To test for the homogeneity of variances of the samples, Levene's test was used.
As can be seen from Table 4, the resulting p-values were greater than 0.05 showing that the variances of the samples were not statistically significant. Consequently, the t-tests are based on equal variances assumed. It can be seen from Table 4 that there were statistically significant differences between the Abagusii students and Luo students in favor of Abagusii students in terms of Source of scientific knowledge [t(408) = 4.466, p < 0.05], Certainty of scientific knowledge [t(408) = 6.461, p < 0.05], and Development of scientific knowledge [t(408 = 2.724, p < 0.05)] as also evidenced by the mean differences.
On the other hand, there were statistically significant differences between the Luo students and Abagusii students in favor of Luo students in terms of justification of scientific knowledge [t(408) = −3.930, p < 0.05] as also evidenced by the p value and mean difference. The effect sizes indicate that 43.1% of the variance in source of scientific knowledge, 60.8% of variance in certainty of scientific knowledge, 26.6% of variance in development of scientific knowledge and 38.1% of variance in justification of scientific knowledge was caused by cultural orientation of the students.

Gender Differences in Science Epistemological Beliefs
To establish whether there were gender differences in science epistemological beliefs, descriptive statistics are first presented followed by inferential statistics. Table 5 below shows the descriptive statistics. Table 5 shows that males had higher scores than females in all the subscales except in the scale of source. However to establish whether these differences are statistically significant, independent sample t test was carried out. Table 6 shows the output of Levene's test and independent sample t-test for the various subscales of SEBQ.
To test for the homogeneity of variances of the samples, Levene's test was used. As can be seen from Table 6, the resulting p-values were greater than 0.05 showing that the variances of the samples were not statistically significant. Consequently, the t-tests are based on equal variances assumed.
It can be seen from the table that there were no statistically significant gender

Discussions
The finding on the effect of "culture" on the development of science epistemological beliefs has indicated statistically significant cultural differences in favour of the Abagusii students for the dimensions source, certainty and justification and statistically significant cultural differences in favour of Luo students for the dimension of development of scientific knowledge. This finding indicates that the Abagusii students had more sophisticated views as regards the source of scientific knowledge as not residing in external authorities (Source). They had more sophisticated views on the fact that a right answer does not exist (Certainty). Finally, they also had more sophisticated views that scientific knowledge evolving and is not static (Development). The Luo students on the other hand hold more sophisticated views on the need to justify scientific knowledge through empirical investigation (Justification). These views could be generally a reflection of the learning environments which the students are presently engaged in. There is a likelihood that the learners are exposed to an environment where scientific knowledge is perceived as tentative, evolving or socially constructed. The findings on Kisii students could also reflect an evolving cultural environment in which the learners believe that knowledge does not exist in external authorities, there is no right answer and that knowledge is evolving. This contradicts the Kisii culture which emphasized recognition and respect of external authorities in the past (Ndeda, 2019). The current study confirms the findings in the international scene that indicated cultural differences for instance, Tabak & Weinstock (2008) who found that Bedouins had higher percentages of absolutist views of knowledge (Certainty) than the Jews in all the domains. Hofer (2007) found out that USA college students were more sophisticated than Japan students in terms of certainty, simplicity, source and justification. Lin et al. (2013) found out that the Taiwanese students generally were more prone to believe that scientific reality is invented, the development of scientific knowledge is always changing and its status is tentative than the Chinese students were. Zhu, Valcke, & Schelleus (2008) found out that compared to the Flemish students, the Chinese students had a higher tendency to believe to a greater extent in "certain knowledge" and to a lesser extent in "fixed ability to learn". Chen and Pajares (2010) found out that Hispanic students ascribe to more naïve beliefs about certainty of scientific know- contradicts the findings some of the findings of Karabenick and Moosa (2005) and agrees with some of them. They found that Omani males were more likely than females to believe that knowledge was simple and originated from scientific authorities. In the same study, there were no gender differences among the US students. The current investigation also agrees with some of the findings of Conley et al. (2004) who revealed that boys and girls were not different in terms of their thinking about source of knowledge, certainty of knowledge or development and justification of knowledge. The current study bears some similarities to some studies for instance Peer (2005) who found that the mean scores of the female students were slightly higher than their male counterparts however the differences were not statistically significant in the four dimensions of epistemological beliefs. The current study departs from the following studies: Chen (2012) found that females tended to be more present in fixed/sophisticated group than males. Cano (2005) found that Spanish secondary male students' beliefs about quick learning were more unstable throughout secondary education. Girls in general expressed more realistic and elaborate epistemological beliefs than boys. Marzooghi et al. (2008), found out that males had more naïve epistemological beliefs than females. Kampa et al. (2016) found out that more females were in the sophisticated and slightly sophisticated dimension whereas more males were in the multiplistic and evidence-based dynamic group. Yenice (2015) found out that gender does not have a significant effect on participants' epistemological beliefs about the role of attempt in learning and there being one truth. However, gender significantly affected the participants' epistemological beliefs on the role of ability in learning in favour of boys.

Conclusion and Recommendations
Based on the findings of the study, the following conclusions are made:  Culture influences the development of science epistemological beliefs: To a larger extent, there are no gender differences in the development of science epistemological beliefs. To a lesser extent, there are gender differences in some dimensions.  In terms of practice, the study recommends that teachers create learning environments where students collaboratively construct scientific knowledge. This will bolster the students' ability to develop more sophisticated epistemological beliefs in science associated with ability to explain and apply scientific knowledge to novel situations. This is likely to minimise the observed cultural differences since epistemological beliefs are modifiable through exposure to progressive learning environments that involve the learners (Weinberg, 2014).

R. O. Ongowo
 There is also need to consistently create gender inclusive pedagogic practices in the classroom so as not to handicap either gender's engagement in learning and performance.
In terms of research, the study recommends the following:  That more within country ethnic based studies on science epistemological beliefs need to be done to build an advanced picture of science epistemological beliefs from the standpoint of culture especially in Kenya which is multi-cultural in nature.  More studies that relate to gender on this subject need to be carried out to expand and conclusively establish the status of gender with regard to development of science epistemological beliefs. There is also need to do more studies from a mixed method perspective in which the weakness of one method is complemented by another.

Limitations
The study has the following limitations: 1) Purposive sampling was employed in this study which reduces the generalization of the findings to a wider population; 2) The sample was also limited to co-educational secondary schools within the two counties.

Conflicts of Interest
The author declares no external sources of funding and also declares no conflict of interest.