Kankam boadu
Download 2.03 Mb.
|
- Navigate this page:
- Sampling procedure
- Sample size
- Samples for the research
PopulationA population is a set of all elements. It is the large group to which a researcher wants to generalize his or her results (Johnson & Christensen, 2008). In this study, the target population comprises all tutors and students of the public colleges of Education in Ghana. In all, there are 38 public colleges of education running the Diploma in Basic Education programme in accordance with the educational reforms in Ghana (Anamuah-Mensah, 2008). There are three forms of colleges of education running basic education in the colleges of education programme in Ghana. These are the general teacher education programme, the specialist (Science, Mathematics, Technical, Early Childhood and French) programmes and the Distance Education programme. The distance education programme has three categories such as the four-year Untrained Teachers Diploma in basic education, the Three-year Diploma in Basic Education for serving certificate “A” teachers and Two-Year Diploma (sandwich) for serving certificate “A” teachers. The accessible population consisted of the final year social studies teacher trainees and the social studies tutors of the colleges of education. The final year teacher trainees of the general education programme were chosen on the following grounds: the social studies students on the general programme spend most of their time on campus and therefore would get time to study concepts and theories of citizenship education and practise them both on campus and in the wider community. The final year teacher trainees were chosen with the view that they might have covered many areas of the social studies syllabus and therefore would be better informed about issues in social studies in general and citizenship education in particular. In all, the total population for the final year teacher trainees was 9,196 with the tutors’ population of 76 in all the colleges of education (Institute of Education, 2009). The distribution of the population of the 38 colleges of education in Ghana is shown in Appendix C.
Sampling can be divided into sample type and sample size. Sample type can be put into two main divisions such as random samples and non-random samples (Borg & Gall, 1989; Cohen, Manion & Morrison, 2004). Walliam (2005) mentions that the benefits of random samples exceed those of the non-random samples in particular, when one is making generalization. He explains that random sampling techniques give the most reliable representation of the whole population while non-random techniques, relying on the judgment of the researcher or an accident cannot generally be used to make generalization about the whole population. Examples of the random sampling techniques are simple random, stratified, cluster, systematic and multi-stage sampling whilst the non-sampling techniques are convenience, purposive, quota, dimensional and quota (Cohen, Manion & Morrison, 2004) Sample sizeSample size determination is inevitable in a research design. The decision on sample size is that the larger the sample size, the better, because larger samples result in smaller sampling errors (Johnson & Christensen, 2008). The mean and standard deviation of a large sample, as Borg and Gall (1989) stress, is more likely to be representative of the mean and standard deviation of the population. Additionally, a larger sample is less likely to lead to negative results (Almaomari, 2007). There is no doubt that using a large sample is acceptable; the researcher, must however, stick to the minimum standard of sample size. Borg and Gall (1989) argue that in survey research, there should be at least 100 participants in each major subgroup, and 20 to 50 in each minor subgroup. The minimum participants in descriptive studies, correlation studies, experimental and comparative studies according to Fraenkel and Wallen (2006), should be 100, 50 and 30 respectively. Samples for the researchThe study is a national survey and in order to give the sample a national representation, attempts were made through appropriate sampling procedures to ensure that teacher trainees and tutors from different colleges of education in Ghana were captured. To achieve this, colleges of education in Ghana were clustered into three ecological zones such as Southern zone, comprising Western Region, Central Region, Greater Accra Region and Volta Region; Middle zone, made up of Ashanti Region, Eastern Region and Brong- Ahafo Region; and Northern zone consisting of Northern Region, Upper West Region, and Upper East Region. Through simple random sampling (lottery approach), one region was selected from each of the three zones. Thus, the Ashanti, Central and Northern Regions were selected for the study. In selecting the teacher trainees and tutors for the study, multi-stage stratified random procedure was used to ensure that the sex –types (mixed and single-sex) of colleges in each region were captured. For each region, two mixed colleges and a single sex college were randomly selected. This was to ensure that the views of all categories of students were represented. In the case of the Northern zone, however, only two mixed colleges were selected to participate because there was no single sex college. This resulted in the selection of eight colleges. After selecting the regions and the sex-type colleges, the colleges were further stratified into years of establishment, where the oldest and the newest colleges of education were purposively selected. This was to make provision for both the new and old colleges’ views to be captured. The procedural directions followed in arriving at the sampled colleges by zones, sex-types and years of establishment are shown in Table 1.
In deciding on the number of teacher trainees to select from each of the colleges, the table for determining the sample size from a given population by Krejcie and Morgan, as cited in Cohen et al. (2004) was used. Krejcie and Morgan (1970) postulate that “as the population increases the sample size increases, at a diminishing rate and remains constant at slightly more than 380 cases” (Krejcie & Morgan, 1970: 610). Borg and Gall (1979: 195) suggest that as a general rule, sample sizes should be large where: there are many variables. only small differences or small relationships are expected or predicted. the sample will be broken down into subgroups. the sample is heterogeneous in terms of the variable under study. reliable measures of the dependents variables are unavailable. The target population for the teacher trainees offering social studies in the colleges selected from the three zones in Ghana was 2,211 (Institute of Education, 2009) . The estimated population of the tutors handling social studies in the selected colleges of education was 36. Krejcie and Morgan (1970) posit that for a population of 2200, a sample size of 327 is quite representative. In eight sampled colleges, their population was given as 2,211. With this figure, it was statistically prudent to use the sample size of 332 in line with Krejcie and Morgan’s (1970) suggestion. This informed the choice of the sample size. Ordinarily, the 332 should have been proportionately distributed among the eight colleges based on their representativeness on their total population. However, this should have led to some colleges not having the minimum of 30 required from a sample to be regarded as normally distributed according to the central limit theory. To avert that problem, 15 percent was selected from each sub population. With the mixed colleges, after selecting the sample, equal proportion (15%) of males and females were selected through the lottery method. In the case of the tutors’ population, all of them (36) were selected because they were not many hence the census method was used. The distribution of the population and the sample of the student-teachers are shown in Table 2. Table 2: Distribution of population and sample selected for teacher trainees
Note. Institute of Education, University of Cape Coast (2009) For the purposes of interviewing respondents for a deeper understanding of issues, in each college, the Heads of Department of Social Studies, and two class leaders of social studies teacher trainees were selected. These respondents were leaders who might have experienced some form of citizenship education as leaders of the colleges. Directory: jspui -> bitstream -> 123456789 123456789 -> College day annual report 123456789 -> A. gw student and alumni numbers summary 3 123456789 -> Clustering Microarray Data within Amorphous Computing Paradigm and Growing Neural Gas Algorithm 123456789 -> Навчальний посібник Для студентів економічних І правових спеціальностей немовних вузів Суми двнз "уабс нбу" 2014 123456789 -> Министерство сельского хозяйства и продовольствия республики беларусь 123456789 -> Personal informations: Nationality: Sudanese Date and place of birth 123456789 -> So far, administration of Koutthep village-fund group (Phonnady) is worked by boards, it has a common coordination and discussion Download 2.03 Mb. Share with your friends:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||