1. Introduction

Self-regulated learning is a process that assists students in managing their thoughts, behaviours, and emotions in order to successfully navigate their learning experiences (Zimmerman, 2006). According to Cook (2014), self-regulation is the ability to monitor and control our own behaviour, emotions, or thoughts, altering them in accordance with the demands of the situation which includes the abilities to inhibit fast responses, to resist interference from irrelevant stimulation, and to persist on relevant tasks even when we do not enjoy them. Aspects of self-regulation correlate with various positive outcomes for students and adolescents including better academic performance, problem-solving skills, and reading comprehension; more satisfying interactions with peers; higher levels of intrinsic motivation, self-worth, perceived competence, self-efficacy, moral cognition, and moral conduct; fewer behaviour problems; and lower levels of psychopathology (Zimmerman, 2006). According to Kochanska, Coy, and Murray (2007), self-regulation is not simply an internal characteristic, but looks at students’ compliance with different requests under four different conditions. The study examined longitudinally the development of self-regulation in 108 young children during the first 4 years of life and found that, it was easier for students to comply when they agreed with the mother's request (for example, they happily helped pick up the toys and continued without being reminded; they said things like "no-no toys" or spontaneously turned away from the prohibited toys) than when they did not agree (for example, they only helped pick up toys when their mothers insisted; they hovered near and touched the prohibited toys.

One of the self-regulated learning techniques studied is task value, which refers to imaginable or objectively measured worth of a task for completion project. For the students under study, tasks included studying, doing assignments, and homework relating to learning. According to Schmid, Fries and Kuhnle (2010) and Nasiriyan et al. (2011), task value can be defined as an imaginable or objectively measured worth of a task for completion project. For the students under study, tasks included studying, doing assignments, and homework relating to learning. Task value is identified by Schmid, Fries and Kuhnle (2010) as an important variable towards predicting or determining academic performances of students. Similarly, Nasiriyan et al. (2011) and Al-Harthy and Aldhafri (2014) noted that task values coupled with self-efficacy are some of the variable combinations that would easily predict academic performances of students.

The Person-centred therapy (PCT) developed by Rogers (1942; 1957) provides clients with an opportunity to develop a sense of self, where they can realize how their attitudes, feelings and behaviour are being negatively affected. Rogers (1942; 1957) affirmed individual personal experience as the basis and standard for living and therapeutic effect. Rogers identified six conditions which are needed to produce personality changes in clients: relationship, vulnerability to anxiety, genuineness, the client's perception of the therapist's genuineness, the therapist's unconditional positive regard for the client, and accurate empathy (Rogers, 1942). This emphasis contrasts with the dispassionate position which may be intended in other therapies, particularly the more extreme behavioural therapies. Living in the present rather than the past or future, with organismic trust, naturalistic faith in your own thoughts and the accuracy in your feelings, and a responsible acknowledgment of your freedom, with a view toward participating fully in our world, contributing to other peoples' lives, are hallmarks of Roger's PCT (Rogers, 1942).

Literature on task value and academic achievement exists. A study by Al-Harthy and Aldhafri (2014) examined whether task value can be used as a good predictor for self-efficacy of students and consequently their academic performances through administering task value and self-efficacy questionnaire to a total of 284 students that were enrolled in different courses within Sultan Qaboos University. The researchers also obtained grade point average (GPA) for the students. The study established that there is a significant relationship between task value and self-efficacy. In addition, the two variables were found to be positively correlated to the GPA of the students. Findings by Al-Harthy and Aldhafri (2014) were consistent with those of Bong (2001), Lawanto (2014), and Chow, Aro and Eccles (2012) who also established the fact that task values are usually positively correlated with the academic performances of students. The researchers did not provide a qualitative exploration explaining how the former affects the latter as it was in the case of current study. The present study sought to conduct a quantitative and qualitative analysis and exploration on the relationship between task values and academic performances of orphaned secondary school students in Kenya.

In another study, Nasiriyan, Azar, Noruzy, and Dalvand (2011) investigated the influence of self-efficacy, achievement goals, task values, and effort on the achievement of students in respect to mathematics. Nasiriyan et al. (2011) employed the concept of path analysis model. The findings by Nasiriyan et al. (2011) indicated that self-efficacy has a direct effect on mastery, performance approach, performance avoidance, and achievement of mathematics. On the other hand, Nasiriyan et al. (2011) established that task value also exert direct effect on mastery goals and efforts. Whereas findings established a negative direct influence of performance approach goals on mathematics achievements, mastery goals were found to be exerting positive effect on effort. The findings of Nasiriyan et al. (2011) are very information especially with respect to establishing the relationship between task value and mathematics achievement. The findings, which are also consistent with those of Lawanto (2014), Chow, Aro and Eccles (2012), and Schmid, Fries and Kuhnle (2010) imply that practitioners in education sector should always focus on enhancing task values of students in order to achieve good mathematics performances amongst students. However, the study was mainly quantitative unlike in the case of the current study in which mixed research was used considering its strength in counterbalancing the shortcomings innate to utilizing separate qualitative and quantitative approaches.

Similarly, Bong (2001) conducted a study to assess the constructs of self-efficacy and task value beliefs through a longitudinal study consolidating self-efficacy and expectancy value theories. The study used a total of 168 Korean female college students in assessing self-efficacy perceptions at varying levels of measurement specificity. On the basis of the exploratory factor analyses, researcher obtained that self-efficacy items were reliably differentiated into separate factors of a priori specificity. Bong (2001) established in the study that all the self-efficacy factors were positively correlated amongst themselves. In addition, the study established that there is a stronger links of self-efficacy to performance and of task value to intentions. However, the study by Bong (2001) looked at two factors, namely, self-efficacy and task value. Bong (2001) focused more on self-efficacy at the expense of task value. Even though the study was longitudinal in nature, the application of qualitative aspects did not focus more on task value unlike in the current study where both qualitative and quantitative methods were used. In addition, a sample of 168 students is smaller compared to the 300 sample that was used in the current study. The study concurs with Lawanto (2014) who reported on how students' perception of course material in terms of importance, utility, and interest is related to their self-regulated learning (SRL) skills and project performance in a web-intensive undergraduate learning environment. The data from 57 students were analyzed. Data sources included survey instruments, ranking questions, and project grades. The research highlights important components of online education by evaluating the connections between students' perceptions of web-intensive course value, SRL, and project performance. Findings show a significant positive relationship between task value and performance, specifically between importance of the activity and performance. However, a sample of 57 students is very small especially when it comes to obtaining data that can be used to replicate educational aspects in a nation compared to the 300 students used in the current research.

According to Chow, Aro and Eccles (2012), two independent studies were conducted to extend previous research by examining the associations between task value priority patterns across school subjects and aspirations toward the physical and information technology- (IT-) related sciences. Study 1 measured task values of a sample of 10th graders in the United States (N 249). Regression analyses indicated that task value group membership significantly predicted sub-sequent aspirations toward physical and IT-related sciences measured 1–2 years later. The reviewed study used a sample of 249 whereas the current study used 300 orphaned secondary school students, which provided more data and information than the reviewed study. In addition, Chow, Aro and Eccles (2012) performed a quantitative study, which meant that a number of qualitative components of the study were left behind; hence, the use of mixed research approach in the current study. Similarly, Gwen, Bohle, Hebert and Wim (2012) analyzed the role of emotions in a virtual world (Second Life) through students' level of enjoyment and boredom and their influence on students' achievement level. The virtual world was an educational tool used to fully immerse students in the content of the course. Pekrun's Academic Emotions Questionnaire (AEQ) was used to measure two academic emotions: boredom and enjoyment. Results from their study show that task value was unrelated to academic performance. However, the reviewed study suffered from weaknesses of using only one research approach unlike in the current study where both qualitative and quantitative studies were adopted.

In another study, Hofer, Schmid, Fries and Kuhnle (2010) investigated the reciprocal relationship between personal value orientations and the experience of motivational interference during studying and leisure. A total of 363 pupils (sixth to eighth graders at the time of first measurement) completed the same questionnaire twice in a 2-year interval. Pupils’ achievement value orientations were connected to differential changes in experiencing motivational interference during leisure and during studying. The reviewed study used 6^th to 8^th graders who are still developing in terms of task value skills unlike in the current study where secondary school students were used, who are assumed to be having a slight improvement on their task value skills. In contrast, Taura (2013) investigated the relationships between self-efficacy, task value, and active procrastination; and what role does self-regulation strategies play in these relationships. A total of 426 pre-service teachers, 223 males and 203 females, studying in colleges of education in North-Western Nigeria participated in the study. The results revealed that task value beliefs were not directly significant predictors of active procrastination. However, the reviewed study targeted a total of 426 pre-service teachers, 223 males and 203 females, studying in colleges of education in North-Western Nigeria participated in the study. The study by Taura (2013) mainly focused on quantitative aspects of the relationship between task value and academic performances. There was no qualitative exploration of the concepts. In other words, Taura (2013) failed to explain in detailed how and why task value should be treated as one of the variables affecting academic performances.

In addition, Nasiriyan, Azar, Noruzy, Ali and Mohammad (2011) from South Africa investigated the influence of self-efficacy, achievement goals, task value and effort on students' mathematics achievement in a path analysis model. Results revealed that task value exerted direct effect on mastery goals and effort. Mastery goals exerted a positive effect on effort but the direct influence of performance-approach and performance-avoidance goals on effort was not significant. The direct influence of effort on mathematics achievement was positive. On the whole, data strongly supported the model indicated that motivational believes are important for their academic performance. The study specifically focused on the influence of self-efficacy, achievement goals, task value and effort on students' mathematics achievement in a path analysis model but not on the influence of self-efficacy, achievement goals, task value and effort on orphaned students' performance and therefore the present study focused on the influence of self-efficacy, achievement goals, task value and effort on orphaned students' performance in secondary schools.

Similarly, Ogundokun and Adeyemo (2013) examined the moderating influence of emotional intelligence, age and academic motivation on academic performance of secondary school students and adopted a survey research design. The participants were 1563 (male = 826, female = 737) secondary school students from Oyo state, Nigeria. Their age ranged between12 years and 17 years with mean age of 15.96 years. Two valid and reliable instruments were used to assess emotional intelligence and academic motivation while achievement tests on English Language and Mathematics were used as a measure of academic performance. Descriptive statistics, Pearson’s product moment correlation and hierarchical regression were used to analyze the data. The results showed that all the predictors were highly meaningful. However, the study picked on two subjects, English Language and Mathematics, unlike in the current study where all the required seven subjects as per the KCSE curriculum were averaged and used.

Ifedayo and Adeolu (2012) in another study examined the relationship between teachers’ performance of instructional tasks and students’ academic performance in the teaching-learning process of secondary schools. The study employed the descriptive survey design and respondents consisted of 60 principals, 540 teachers and 1800 students that were randomly selected using the multi-stage sampling technique from a pool of 599 public secondary schools (now re-articulated into 301) in Ondo State, South-West, Nigeria. Results show that the relationship between teachers’ instructional tasks and students’ academic performance is significant (r = 0.828 at p < 0.05) with the teachers’ tasks (Mean = 42.17) and students’ performance (Mean = 2.73). The reviewed study used descriptive survey design, which lacked qualitative dimensions while the current study used mixed method design. The reviewed study was carried out on students in general whereas the present study was done specifically on orphans who are considered special groups of students considering a number of challenges that they are likely to encounter throughout their learning.

In another study, Mucherah and Stahl (2014) investigated the relationship between middle school students’ reading motivation and achievement originates primarily in the Western world. The study was conducted in Kenya and the United States to investigate whether motivation to read and reading achievement are related similarly in the 2 settings. A total of 208 seventh-grade students from public schools from Kenya (51 girls, 34 boys) and the United States (60 girls, 63 boys) completed the Reading Motivation Questionnaire Wigfield & Guthrie, (1995) after completing their respective standardized examination for reading achievement. Regression analyses revealed that task value was predictors of reading achievement for the U.S. students while challenge and compliance were both positive significant predictors of reading achievement for the Kenyan students between task value and academic performance among orphaned students in secondary schools. The sample size used for Kenya was only 123 compared to the current study that used 300 sample, which means that more information was collected in the current study than former.

On a different perspective, Matemba (2014) investigated the relationship between problem solving approach and academic performance among secondary school students in Kakamega Municipality. Correlation research design was adopted for the study. Using stratified sampling, six schools consisting of one boy’s school, one girl’s school, three public co-educational schools and one private co-educational school were sampled. Proportionate and simple random sampling was used to select a total of 200 students comprising of 113 boys and 87 girls. Questionnaires, observation check lists and Focus Group discussions were used to collect data. Quantitative data collected in the field was analyzed using descriptive and inferential statistics, while qualitative data from Focus Group Discussion was analyzed qualitatively. There was no significant relationship between problem solving approaches and academic performance. However, apart from establishing no significant relationship between problem solving and academic performance, which the current study aimed at finding out if it is true, the reviewed study used a sample of only 200 unlike in the current study that 300 sample was used.

Similarly, Waweru and Nyagosia (2012) examined the relationship between utilization of allocated time on task and academic performance of secondary school students in Kenya. The study employed the survey design to obtain the research data from secondary schools in five Counties in Central Kenya. Stratified sampling was used to select 40 schools, comprising 20 schools from the top performing category and 20 from the bottom performing category. Study participants comprised 381 respondents drawn from the 40 schools. Data was collected using questionnaires and analyzed using Pearson correlation coefficient. The study found that schools performing well in national examinations were putting more emphasis on time-on-task. The reviewed study was purely quantitative, which did not give room to explain why and how unlike in the case of the present study where both quantitative and qualitative designs were adopted.

In another study, Ibrahim (2014) investigated whether task value can predict students’ self-efficacy and how the two variables relate to students’ academic performance. A task value and self-efficacy questionnaire was administered to 284 students enrolled in different courses at Sultan Qaboos University (SQU). Grade Point Average (GPA) data from these randomly selected students were collected and the analysis was done using Pearson correlation coefficients, independent sample t-test and regression analyses. Findings realized a relationship between task value and self-efficacy and both variables significantly correlated with the students’ GPA. This study was carried out on university students while the current study was done on orphaned secondary school students given that they have varied task value orientation with the university students specializing in few areas whereas secondary students working on many subjects at the same time.

Similarly, Barkley (2006) examined whether sixth, seventh, and eighth grade students’ efficacy beliefs had any influence on reading comprehension subtest score on the Stanford Achievement Test. The results of this study revealed a positive correlation between subjects’ efficacy beliefs in regards to prior knowledge, self-monitoring and graphic organizers and their reading comprehension achievement. Substantial amount of research support the conclusion that task value is positively correlated to academic performance (Alharthy et al., 2010; Metallidou & Vlachou, 2010). In addition, students with high task value were found to be more cognitively engaged. In a study conducted by Metallidou and Vlachou (2010) looked at the relationship between teachers’ evaluation of 5^th and 6^th graders and students’ task value. The findings of a task value questionnaire showed that students with high task value beliefs in Mathematics were described as more cognitively, meta-cognitively, and motivationally competent learners, as compared to students with lower task value beliefs. The reviewed study was mainly quantitative unlike in the current study where mixed research approach was used to help in counterbalancing the shortcomings innate to utilizing each approach separately.

In another study, Liem et al. (2008) investigated the relationship between English achievement, task value and self-efficacy. The findings showed that prior English achievement correlated with task value and self-efficacy. Self-efficacy and task value predicted students’ goal orientation, with task value being the more predictive of the two for mastery goal adoption. However, Liem et al. (2008) focused mainly on quantitative aspects of the relationships unlike the present study that adopted both qualitative and quantitative approaches for purposes of looking at both the numerical and non-numerical analysis besides counterbalancing the shortcomings that one approach may have. The qualitative approach’s weaknesses were counterbalanced by the strengths of quantitative approach in the current study. In order to establish the relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-Country, the following hypothesis was tested using Pearson Product-Moment correlation.

H₀: There is no statistically significant relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-County, Kenya.

2. Research Methodology

The current study used a concurrent quantitative-qualitative explanatory method where the qualitative results were used to help in explaining the quantitative results. The data streams were integrated first to help select the interview participants and to refine the interview questions and secondly once both data streams have been analyzed (Creswell, 2014). This design was relevant to this study it enabled the researcher explore in details issues of self-regulation and how they play a role in the academic performance of the orphan (Creswell, 2014). The 35 head teachers and 300 orphaned secondary school orphans in Bondo Sub-County were targeted as respondents in this study (CDEO’s report 2015). The target population was form four orphaned secondary school students in Bondo Sub-County from which representative sample was obtained. The reason for focusing on form four students is because based on the present curriculum, the academic performance of a student for four years is determined at the end of form four where all that has been learned through the years is tested. The sample size constituted 300 orphaned secondary school students and 11 head teachers. This study picked eleven schools from which the 11 head teachers were used as respondents. This was based on recommendation by Jwan (2010) stating that in any research with a known population 10% to 30% of the target population should be considered as appropriate. In this case, the study picked 30% of 35 schools, which are 10.5 ≈ 11 h In Part V, the students’ questionnaire contained items relating to task value skills. Task value is an imaginable or sometimes objectively measured worth of a task for completion of a project whereto it belongs (Chow, Aro & Eccles, 2012).

Task value usually serves as a basis for prioritization of project tasks, so the most impactful activities can be supplied with the best resources and hence to be done at the first place (Nasiriyan, et. al., 2011). The items adopted the Likert scale method whereby respondents had to respond to every statement using a structured format: Strongly Agree (SA), Agree (A), Not Sure (NS), Disagree (D) and Strongly Disagree (SD). To measure academic performance, document analysis of mock results of the orphans 2015 was used (Huertas et al., 2015) and identified the mean grades of every participant (Appendix V). Document analysis is a form of qualitative research in which documents are interpreted by the researcher to give voice and meaning around an assessment topic (Bowen, 2009). Analyzing documents incorporates coding content into themes similar to how focus group or interview transcripts are analyzed (Bowen, 2009). A rubric can also be used to grade or score document. In this case a mean grade of A was represented by 12 points whereas a mean grade of E was represented by 1 point. The reason for using mean grades for the MOCK exams is because it represented the academic performance of the participants in all subjects.

In this study, document analysis was considered as efficient and effective in gathering data because documents reviewed were manageable and practical resources. Documents are commonplace and come in a variety of forms, making documents a very accessible and reliable source of data. Obtaining and analysing documents was far more cost efficient and time efficient than conducting other research or experiments on performances of the orphaned secondary school students (Bowen, 2009). Documents reviewed were considered stable, “non-reactive” data sources, meaning that they can be read and reviewed multiple times and remain unchanged by the researcher’s influence or research process (Bowen, 2009, p. 31). Document analysis was used because of the many different ways it supported and strengthened research. Document analysis was also used in providing background information and broad coverage of data, and therefore helpful in contextualizing the research within its subject or field (Bowen, 2009). The present study conducted qualitative in-depth interviews with head teachers who had a picture of the performances of different sets of students including orphans who were the study subject (Kombo & Tromp, 2006).

To determine the validity of the instruments, a pilot study was conducted in two schools from the neighbouring Rarieda Sub-County. The questionnaires were issued to the 75 students that formed the pilot study sample while the interview schedule was administered to 8 head teachers that formed the piloting sample. On the other hand, the internal consistency of the scale yielded a Cronbach’s alpha of .891 from all the items that were included within the general questionnaire. The researcher obtained a letter of permission from the director Board of Post Graduate Studies of Jaramogi Oginga Odinga University of Science and Technology (JOOUST) (Appendix VIII). Additionally, the researcher obtained a permit of permission to conduct the research from the National Council of Science and Technology for Innovation (NACOSTI) (Appendix IX) and consequently clearance from the county.

A combination of descriptive and inferential statistical analyses were used. Descriptive statistical analysis was done using frequencies and percentages while the inferential statistical analysis involved using Pearson correlation and regression analysis. The academic performance was also reported for each student and correlation determined between the various independent variables and academic performance. Hypotheses were tested at the 5% level of significance (p = 0.05). Null hypothesis was rejected when p-value < 0.05 and accepted when p-value > 0.05. Data from interviews were analyzed using thematic analysis which followed the principles of thematic analysis (Braun and Clarke, 2006).

3. Findings & Discussion

In order to establish the relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-Country, the following hypothesis was tested using Pearson Product-Moment correlation.

H₀₄: There is no statistically significant relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-County.

H_a4: There is statistically significant relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-County.

Rule for hypothesis testing states that the p-value of 0.05 was considered to either reject or accept the null hypothesis. In this respect, when the p-value ≤ (less than or equal to) 0.05, the null hypothesis was rejected and alternative hypothesis accepted instead. On the other hand, if the p-value > (greater than) 0.05, then the null hypothesis is accepted and alternative hypothesis rejected. The results of Pearson Product-Moment correlation testing are summarized in the following table:

Table 1. Correlations between Task Value and Academic performance

Results in Table 1, p < 0.05, which indicates that the null hypothesis is rejected confirming that the correlation is statistically significant (r = .122, p < 0.05). The Pearson Product-Moment correlation coefficient (r = .122 < 0.5) computed indicated that there was weak (Heeringa, West, & Berglund, 2010) positive correlation between task value learning strategies and academic performance among orphaned secondary school students in Bondo Sub-County. Since the p-value obtained is 0.004 < 0.05, the null hypothesis stating that “there is no statistically significant relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-County” is rejected and alternative hypothesis stating that there is statistically significant weak positive relationship between self-efficacy beliefs and academic performance among orphaned secondary school students in Bondo Sub-County is accepted.

The Pearson Product-Moment correlation coefficient (r = .122) computed indicated that there was weak positive correlation between task value and academic performance among orphaned secondary school students in Bondo Sub-County. The analysis revealed highly significant (p < 0.05) positive though weak relationship between task value and academic performance, with high levels of task value associated with the fact that majority of the students enjoy doing the subject material offered in school, some considered learning activity as instrumental in meeting their long- or short-range goals. The findings concur with the findings by Lawanto (2014) who showed a significant positive relationship between task value and performance, specifically between importance of the activity and performance. Further, results showed a positive correlation between goal setting and performance. Significant positive correlations were also found between task value and goal setting, task strategies, help seeking, and self-evaluation. In addition, the present findings are also consistent with those of Bong (2001), Nasiriyan et al. (2011), and Al-Harthy and Aldhafri (2014). Al-Harthy and Aldhafri (2014) established that there is a significant relationship between task value and self-efficacy. In addition, the two variables were found to be positively correlated to the GPA of the students. Nasiriyan et al. (2011) on the other established that task value also exert direct effect on mastery goals and efforts. Bong (2001) as a good predictor for academic performances, task value is also a good predictor of intentions by the students towards attaining some levels of achievements or performances. Scatter plot for task value and academic performance is presented in figure 1 as follows:

Figure 1. Scatter-plot for Task Value and Academic performance

The scatter plot above implies that all task value variables are around the line of best fit. The other implication of this scatter plot is that both variables move in the same direction. In other words, as one variable increases, the other variable also increases. As one variable decreases, the other variable also decreases. In this case, increasing task value concepts would result into an increase in academic performance. The present study findings are consistent with those of Nasiriyan et al. (2011) and Al-Harthy and Aldhafri (2014) who confirmed that task values are essential when it comes to determining academic performances of students. However, the present study findings are inconsistent with Nasiriyan et al. (2011) and Hassanbeigi et al. (2011), which indicated that there are instances when the relationship between task value and academic performance is non-existence; thus, not guaranteed that when a student has adequate task value skills he or she will perform better.

4. Conclusions & Recommendations

The study established a relationship between task value and academic performance among orphaned secondary school students in Bondo Sub-County, Kenya. Majority of the students engage in various tasks owing to the values attached. From the present study findings, there are various task value skills or skills that orphaned students in secondary schools employ in enhancing their academic performances. However, the study also confirms that task values are not synonymous to some of the orphaned students within secondary schools in Bondo Sub-Country. ii. School guidance and counsellors should review and redesign effective policies relating to self-regulation skills such as meta-cognitive skills, time management skills, self-efficacy skills, and task value skills, which can be used in enhancing the academic performances of special interest group of students such as the orphans.