CHAPTER THREE RESEARCH METHODOLOGY
This chapter focuses on research design, population, sample and sampling methods, data collection instruments, pilot testing, ethical issues, the data collection procedure, and methods of data analysis.
In this study, the relationships between perceived parenting style/dimension, TEI, and university adjustment amongst first year students were examined. The study employed descriptive survey research in which parenting dimensions (parental demandingness and responsiveness) stood as the independent(exogenous) variables, while TEI as the dependent (in the one way ANOVA and Independent sample T-test analysis), the independent (in the case of regression analysis), endogenous variable (in the path and mediation analysis), and presumed mediated variable (in the mediation analysis), and adjustment was considered as the dependent(in the case of regression, one way ANOVA, and independent sample T-test analysis), and endogenous variable (in the case of the path and mediation analysis). A descriptive survey research design was chosen because it is a very appropriate one to examine the relationship between the independent and dependent variables.
In addition, this study employed a quantitative (positivist) approach method. A quantitative research approach was chosen especially over the qualitative (interpretivist) one due to the following reasons. The data in quantitative research is basically used to compare and contrast other research and can be used to create new theories and/or test existing hypotheses, which are the very objective of this research and which cannot be explored using qualitative methodology. Nonetheless, quantitative data has a limitation, amongst others, in that it does not recognise the individuality of participants under normal conditions.
Qualitative research, on the other hand, rejects the notion of there being a simple relationship between our perception of the world and the world itself, instead arguing that each individual assigns different meanings to different events or experiences. Due to the individual, subjective nature of qualitative data, it is often inappropriate or not even possible to make predictions for the wider population. Most importantly, because of the open ended approach used in qualitative research, it may be difficult to test hypotheses of a study like the present one. This research by its nature employed constructs with sub variables in them which cannot be fully explored using a qualitative method; as a consequence, the quantitative approach was chosen over the qualitative method.
Once a decision was made about the method of research to be employed in this thesis (quantitative), defining the population to be used in this thesis for the readers was logical. This study was conducted in public universities in Ethiopia. This country has 43 such higher institutions (MoE, 2014). Of them, only 33 public universities have been functioning while the construction of the remaining 10 universities was not completed so that there were no students registered in them during the period of data collection for this research. Of the remaining 33 universities, three universities (Adama Science and Technology University, Addis Ababa University, and Debre Berhan University) were selected randomly using a lottery method.
Basically, there are several approaches that are aimed to minimize bias in the process of random sampling selection, however, for this research purpose a lottery method was chosen among them. In the selection process of research sites, each of the 33 public universities was assigned using random numbers, between 1 and 33. Thus, each university were numbered systematically and in a subsequent manner by writing each number on a separate piece of paper. These pieces of papers were folded and mixed up into a box and then numbers were drawn out of the box in a random manner. Lastly, three sample universities were selected randomly from the box by choosing folded pieces of papers in a random manner. Since there were no duplicates in numbers, each university was only sampled once i.e. such selection was sampling without replacement.
First year university regular undergraduates were the research population for this study because the investigator specifically wanted to examine the particularly stressful time of university life as these students transit from preparatory school and life at home to life at a university. Moreover, regular first year undergraduates, since they originated from diverse areas of the country, could face more challenges of adjusting to a university than the weekend students who mostly came from the local districts.
SAMPLE AND SAMPLING METHODS
The sampling frame for this study was first year regular undergraduates who had enrolled in the academic year 2016/2017 at 3 public universities of Ethiopia. The samples, from which statistical inferences were drawn, were randomly selected from the accessible populations; thus, generalising the findings from selected samples to the population was feasible. Random sampling assumes that the units to be sampled were included in a list, therefore, list were numbered in sequential order from one to the total number of units in the population.
Concerning selection of colleges and departments, a simple random sampling technique (lottery method) was employed. A simple random sample is a probability sampling method that provides an equal probability of each member of the population being selected. In other words, simple random sampling ensures an unbiased representation of a population under considerations. In a simple random sample, a couple of methods can be used, among them; a lottery method is one and was used in this research. Although a lottery method seems mechanical, in this research context it was applied and used without difficulty. Since the population was manageable, this method was applicable in this sense. In this lottery method, each member of the population (in this research, colleges, departments, and participant’s names who were chosen from the selected departments after stratification was made) was numbered systematically in a consequent manner and writing each number on a separate piece of paper of same size, shape, and colour followed. Then after, those pieces of papers were folded and mixed up in a box and lastly, samples were drawn out of the box in a blindfold random manner, until the required samples were taken for granted. Since the random numbers were mutually exclusive, each sample was only sampled at once or draw a sample, without replacement, means that once an individual was sampled, that sample was not placed back in the population for re-sampling.
In this regard, in the Adama Science and Technology University, there were two divisions in the first-year programmes during the data collection period. These were the Pre-engineering freshman programme and the Applied Natural Science division. From these two divisions, the Pre-engineering division was selected randomly using a lottery method.
In the case of Addis Ababa University there were 14 colleges and institutions during the data collection period, but the Institute of Engineering was excluded since it had been selected in the Adama Science and Technology University; thus, only 13 colleges and institutions were included in the selection process. Of these, two colleges, the Social Science and Natural Science colleges, were chosen using a simple random sampling technique. Again from the two colleges, 3 departments were chosen randomly using a lottery method. In the Social Science College, there were 9 departments during the data collection period: of them, Geography and Environmental studies, History and Heritage Management, and Sociology were chosen randomly using simple random sampling techniques (a lottery method). There were 9 departments under the Natural Science College; of them, three departments, namely Geology, Sport Science, and Computer Science, were chosen randomly using a lottery method.
Concerning Debre Berhan University, there were 9 Colleges during the data collection period. Of them, three colleges (Pre-engineering, Social Science and Natural Science) were excluded due to being included in the selection process in the two universities (Adama Science & Technology and Addis Ababa University mentioned above). Therefore, out of the 9 colleges six were left to be included in the sample. Thus, two colleges: College of Business and Economics and College of Health Sciences were chosen randomly using a lottery method. In the college of Business and Economics, there were five departments: of them, the Accounting and Finance, Management and Logistic and Supplies Management departments were chosen randomly using a lottery method. Once more, in the College of Health Sciences, out of the six departments, three: Midwifery, Nursing Science, and Health Officer were chosen randomly using a lottery method.
In selecting participants from each of the departments chosen, caution was taken to ensure the proportionality of the number of students to be included. For instance, in some departments there were numerous students enrolled and they contained many sections; therefore, greater numbers of participants were chosen than from a department having lower numbers of students in a section. Moreover, in choosing the colleges from the selected universities, randomisation of the colleges was not carried out haphazardly; rather, caution was taken to avoid double selection of colleges across the three universities. For instance, if an Engineering college had an opportunity to be chosen once at one university, it was excluded from the selection process applied in other universities.
It is obvious that the quality of the sample affects the quality of the research generalisations. Accordingly, obtaining an unbiased sample is the main criterion when evaluating the adequacy of a sample. An unbiased sample is one in which each individual of the population has an equal chance of being selected. In other words, all members of the population have essentially the same probability of being included. A good sample is also comprehensive in nature as well. This feature of a sample is closely linked with true representativeness. Therefore, the probability sampling technique was applied in this research because it was an objective method of sampling and permitted the application of statistical devices as planned. In probability sampling, the error due to sampling can be estimated. This also maintains the accuracy of the analysis of results as compared to the non-probability sampling technique.
Among the probability sampling techniques, a stratified random sampling method was employed in selecting the participants of this study. This technique was chosen because it was believed to adequately represent the subgroups as well as to ensure a proportional number of the population in the sample. Thus, the stratified sampling technique helps to avoid over or under representations of a segment of the population in the subgroups. Overall, the process of selection was carried out in the following manner. First, participants were stratified based on the required demographic variable of sex, since gender was identified as one essential component for the study. Thereafter, the required number of participants was selected from each distinct stratum via a simple random sampling technique using a lottery method. The number of participants to be selected was determined by a proportional method. This proportional stratified random sampling should ensure whether or not the subgroups (in this case male and female first year university students) were represented in the correct proportions. In this regard the same percentage of participants, not the same number of participants, was drawn from each stratum.
Once the probability sampling type was determined, the next step was to establish the number of participants included in the sample. In survey studies, a sample should be representative of the population. Therefore, the size of the sample is an important aspect for representativeness. Basically, in determining the sample size, a number of factors need to be considered, such as population size, margin of error (confidence interval), confidence level, and number of variables used in the research and the statistical analysis technique to be used as well as time, money, and effort.
Generally, the best answer to the question of size is to use as large a sample as possible. A larger sample is much more likely to be representative of the population (such a sample provides greater confidence to the general population). Other things being equal, the larger the sample, the greater the precision and accuracy of the data it provides as well as the smaller the standard error. Despite those facts, sample size alone does not qualify the ability to generalise. A small sample may effectively represent the population, if the participants of the study are selected randomly or if members of the population of the study are accorded an equal chance to be included in the selection. Basically, there is no single rule that can be used to determine sample size. The exact procedure by which to determine the sample size required varies with the nature of the variable and its sampling distribution, but the basic procedure can be illustrated in connection with the mean of random samples based on normal probability distribution.
There are several approaches to determining the sample size: a census for small populations, copying a sample size of similar studies and applying formulas to calculate a sample size.
- The first approach is to use the entire population as the sample. Here, the entire population would have to be sampled in small populations to achieve a desirable level of precision. In this research, this approach would have been highly impractical because the population size was too large to accommodate in this research.
- The second approach is to use the same sample size as those of studies similar to the one the investigator plans for. Without reviewing the procedures employed in those studies, one may run the risk of repeating errors that were made in determining the sample size for another study. For this reason, this approach was not chosen for this study.
- The third approach is to use a formula to determine the sample size. In this regard, Yamane (1967) provides a simplified formula to calculate this size. This formula was used to calculate the said size.
CHAPTER ONE: STATEMENT OF PROBLEM, AIM AND SCOPE OF THE STUDY
1.1 BACKGROUND AND MOTIVATION
1.2 PROBLEM STATEMENT
1.3 AIM OF THE STUDY
1.4 OPERATIONAL DEFINITION OF TERMS
1.5 SIGNIFICANCE OF THE RESEARCH
1.6 ASSUMPTIONS AND SCOPE OF THE RESEARCH
1.7 CHAPTER DIVISION
CHAPTER TWO: REVIEW OF RELATED LITERATURE
2.2 NOTIONS ABOUT EI
2.3 PARENTING STYLE
2.4 RELATIONSHIP BETWEEN MAJOR STUDY VARIABLES
2.5 THE RELATIONSHIP BETWEEN GENDER AND MAJOR STUDY VARIABLES
CHAPTER THREE: RESEARCH METHODOLOGY
3.1 RESEARCH DESIGN
3.3 SAMPLE AND SAMPLING METHODS
3.4 DATA COLLECTION INSTRUMENTS
3.5 ETHICAL CONSIDERATIONS
3.6 DATA COLLECTION PROCEDURE
3.7 METHODS OF DATA ANALYSIS AND STATISTICAL PROCEDURES
CHAPTER FOUR: RESULTS OF THE STUDY
4.1 DEMOGRAPHICS OF THE STUDY SAMPLE
4.2 RESULTS OF THE PRELIMINARY ANALYSIS
4.3 VARIABLES TO BE INCLUDED IN A PATH ANALYSIS AND THE CORRELATION BETWEEN THESE VARIABLES
4.4 MULTIPLE REGRESSION ANALYSIS
4.5 MEDIATION ANALYSIS
CHAPTER FIVE: DISCUSSION
5.1 THE PARENTING STYLES PREDOMINANTLY PRACTICED IN ETHIOPIA
5.2 TEI AS A FUNCTION OF PERCEIVED PARENTING STYLE
5.3 ADJUSTMENT AS A FUNCTION OF PERCEIVED PARENTING STYLE
5.4 DIFFERENCES IN TEI AND ADJUSTMENT BY GENDER OF THE STUDY PARTICIPANTS
5.5 DISCUSSION ON RESULTS OBTAINED FROM THE PATH MODEL, MULTIPLE REGRESSION,AND MEDIATION
CHAPTER SIX: SUMMARY AND CONCLUSIONS OF THE STUDY
6.3 STRENGTHS, LIMITATIONS, AND FUTURE RESEARCH
6.4 PRACTICAL, THEORETICAL, AND METHODOLOGICAL CONTRIBUTIONS OF THE STUDY
6.5 RECOMMENDATIONS FOR FUTURE INTERVENTIONS
GET THE COMPLETE PROJECT
PARENTING STYLE AND FIRST YEAR STUDENTS’ ADJUSTMENT AT UNIVERSITY: MEDIATION VIA TRAIT EMOTIONAL INTELLIGENCE IN HIGHER EDUCATION INSTITUTIONS- A DIMENSIONAL AND TYPOLOGICAL APPROACH