CHAPTER THREE RESEARCH METHODOLOGY AND DATA COLLECTION
This chapter deals with the methods and procedures followed in the collection and interpretation of data. The tools utilised in collecting data include experiments, interviews, questionnaires and textbooks. The sample population is explained and data are recorded and interpreted. The chapter further summarises the results of the data collected.
Research population and sampling
The study made use of participants who would provide the researcher with the required information. Therefore, participants were selected because of their involvement and knowledge of the subject under investigation.
According to Fraenkel, et al. (1993:9), ‘population’ refers to ‘the group to which the researcher would like the results of a study to be generalizable; it includes all individuals with certain specified characteristics’. This means that the items from which the researcher can select subjects for the study should have certain features that can enable the researcher to generalise.
The study consisted of 508 tertiary and high school students who are Sesotho mother-tongue speakers, to test the effectiveness of Sesuto-English Dictionary and Sethantšo sa Sesotho (i.e. 254 students in each case) and 40 language experts. The population of this study thus consisted of 548 mother-tongue speakers of Sesotho, which comprised of 163 males and 385 females altogether. This group was made up of 434 high school students from five districts, namely Mafeteng, Leribe, Mokhotlong, Qacha’s Nek and Quthing. The number of learners varied from place to place due to the number of students that were present in a class during the test. The remaining 74 participants were student teachers and those training to be translators from tertiary institutions, namely the Lesotho College of Education (LCE) and National University of Lesotho (NUL); and 40 language experts (10 teachers, 10 lecturers, 10 people from media houses and 10 Sesotho Academy members).
The researcher chose Mokhotlong, Qacha’s Nek and Quthing districts to represent the people of the highlands while the lowlands are represented by Mafeteng and Leribe including Maseru since LCE, NUL, media houses, lecturers and some members of the Sesotho Academy are located in Maseru.
This was done to ensure that both areas were represented, as this would help the researcher observe the vocabulary that is used in both regions. The researcher assumed that the Sesotho used in the highlands is slightly different from the Sesotho used in the lowlands due to the external influence and rapid changing world, particularly in urban areas. People from the lowlands are exposed to technology and other factors before those from the highlands. Again, foreigners visit the lowlands more often and their languages have influenced Sesotho, i.e. the Sesotho used in the lowlands is somewhat mixed; For instance, the word ‘road’ is called pata, by people from the south, which is part of the highlands, and ‘mila or tsela by those from the north, which is part of the lowlands. This indicates that words that may seem unfamiliar to the lowlands people may be common in the highlands and vice versa.
‘Sampling’ deals with the selection of a group from whom data is obtained. This implies that instead of collecting data from the entire population of interest, the researcher may select only some members of the population. The results obtained from the selected group are used to ‘make generalizations about the entire population only if the sample is truly representative of the population’ (Leedy & Ormrod, 2001:210). There are different types of sampling which are determined by various situations. Leedy and Ormrod (2001:210-219) stipulate that there are two main categories of sampling namely probability sampling and non-probability sampling. Probability sampling consists of simple random sampling, stratified random sampling, proportional stratified sampling, cluster sampling and systematic sampling while non-probability sampling comprises of convenience sampling, quota sampling and purposive sampling, the latter of which is employed in the current study.
Purposive sampling was used for the selection of the sample in this study. According to Babbie and Mouton (2001), in purposive sampling, participants are selected according to the researcher’s judgement about which units are the most useful or representative. This method is also called judgemental sampling. The sample was selected because they had no prior knowledge of what was expected, which suggests that the selected sample were representative or because they had the needed information (Fraenkel, et al. 1993). In addition, Tongco (2007) mentions that samples are selected based on the participants’ knowledge and information about a particular issue. The researcher chose this method, since the study requires information from learners and language practitioners in particular. The Sethantšo sa Sesotho stipulates in its back matter that it is intended to be used by learners from high schools, tertiary institutions and lecturers of the South African Development Communities (SADC). Learners and language practitioners were regarded as being suitable for the study because of their involvement in dictionary usage and their knowledge and information about this issue could contribute to the study.
The sample of the current study was heterogeneous in nature in the sense that it consisted of people whose levels of knowledge of the Sesotho use and experiences were different, i.e. ranging from high to medium to low. High school students were used in this study to represent the low level, student teachers and translation trainees represented the medium level and language experts represented the high level. Each homogenous group was tested separately based on the participants’ level of knowledge, since it was understood that individuals in a group would feel free to engage fully in the discussion if they felt comfortable with each other (Krueger & Casey, 2000; Richardson & Rabiee, 2001). High school students were separated from student teachers / translation trainees and the tests were given in different locations as the groups were situated in different areas. The researcher adhered to the principle that the participants in a focus group should have the same gender group, age-range, ethnic or social class as stipulated by Krueger and Casey (2000). This type of sampling method was found to be appropriate for this study since it ensured that the different groups of population were sufficiently represented in the sample (Nachmias & Nachmias, 1981; Chadwick, et al., 1984; Singleton, et al., 1988).
Question 1: All students were provided with a list of selected words from both the Sethantšo sa Sesotho and the Sesuto-English Dictionary, and were required to use those words in sentences of their own (see Appendix 3).
Question 2: All students were given a reading comprehension exercise. Students were divided into two groups with an equal number of students in each group where possible. One group was allowed to utilise dictionaries to answer questions while the other group did not use dictionaries at all.
Interviews: Unsuspected interviews were conducted using focus groups where the participants’ opinions were sought. The language experts were given questionnaires instead of interviews due to their experience. They were also required to give their views about the two dictionaries and to state their expectations about Sesotho dictionaries in general. Their responses were used for inductive reasoning.
Generalisation was therefore made based on the sample of 548 (i.e 508 learners & 40 language experts). The study also tried to find out the types of words participants might want to see in future dictionaries.
The two dictionaries were compared by covering 19 alphabetical letters, namely a, b, ch, e, f, h, i, j, k, l, m, n, o, p, q, r, s, t and u. These alphabetical letters were used to arrange words in the Sethantšo sa Sesotho. This implies that all the words in the Sethantšo sa Sesotho were used as the sample, while in the Sesuto-English Dictionary, the study utilised only the items that are similar to those contained in Sethantšo sa Sesotho. The following section discusses the procedure and specific data collection techniques used in this study.
Methods of data collection
As was mentioned earlier, this study employed triangulation, as it utilised combined-methods of data collection. The study used various methods of data collection such as experiments, interviews, questionnaires and textbooks. The next section presents such methods.
According to Fraenkel, et al. (1993:4), an experimental study is ‘a research study in which one or more independent variables are systematically varied by the researcher to determine the effects of this variation’. The researcher used tests to investigate a specific problem. For Leedy and Ormrod (2001), experimental study looks at the possible influences that one situation may have on another. Babbie and Mouton (2001) submit that experiments require one to take action and perceive the results of that action. This study utilized experimental study to investigate the effectiveness of the Sesuto-English Dictionary and Sethantšo sa Sesotho in reading and writing as well as dictionary use when learning Sesotho. The experiment was intended to test a group of learners who are Sesotho mother-tongue speakers to determine their performance when making use of the Sethantšo sa Sesotho and the Sesuto-English Dictionary while reading and writing Sesotho. This would assist the researcher, teachers and the students to discover the consequences of dictionary use when learning Sesotho.
During the visits to different schools, the head of the department from each school introduced the researcher to the students and gave her time to explain to the students how the activity would be carried out. The researcher explained to the students that the main purpose of the test was not to award learners marks but rather to investigate how they would perform in their Sesotho lessons when they use dictionaries to read and write Sesotho. The researcher appealed to students to be sincere with their answers, and explained that those who were asked to guess were not to feel bad if they did not know the word, all they had to do was guess the meaning. The researcher further explained that students would be divided into two groups: those who would use the dictionaries and those who would not be using any dictionary. After dividing the learners into two groups, learners were given a chance to choose whether to use a dictionary or not. To the researcher’s surprise most students wanted to answer without the use of the dictionary because they felt no need for it, seeing that Sesotho was not their second language. However, eventually one group decided to use the dictionaries.
Thereafter, the researcher distributed question papers, answer sheets and the copies of the dictionaries, as agreed. The two groups were distanced so that the group who was not utilising the dictionaries could not be tempted to look at the dictionaries used by the other group. Students were requested not to write their names on the answer sheet, but to write the name of the school/institution instead, and to write whether they used the dictionary or not next to the name of the school/institution (i.e. they were to write ‘dictionary’ if they used it or ‘no dictionary’ if they did not use it). Learners were given 40 minutes to finish Question 1 and Question 2.
The test was given to 508 participants (434 high school students and 74 student teachers / translation trainees from the LCE and NUL). The students within each of the two groups were from the same grade or level of study (i.e., Grade 11 high school students and third year students from LCE and NUL respectively). For the first question, all students were provided with a list of selected words from the dictionaries and were required to use those words in sentences of their own, while the second question was a reading comprehension exercise.
Students were divided into two equal groups (as mentioned earlier) where possible. In some instances, the researcher was faced with a challenge of odd numbers where one group consisted of say, 43 students, and in such cases, one sub-group consisted of 21 students while the other sub-group consisted of 22 students. Thus, the total number of students who were not using dictionaries had two students more as opposed to those who used dictionaries (see Table 3.1 below). Their scores were used to judge the effectiveness of both the Sesuto-English Dictionary and Sethantšo sa Sesotho and dictionary use when learning Sesotho as a native language.
The students provided sentences for Question 1 and their responses were classified into three categories, namely correct sentences; wrong sentences, and no answer. The first column shows the location of the particular group; the second one indicates the overall percentage of each group of correct sentences; the third column indicates the percentage of wrong sentences, and the fourth column shows the no response rate for both dictionary users and non-dictionary users respectively. The subsequent table presents the summary of the overall results for Question 1 for both dictionary users and non-dictionary users from seven different groups in the case of Sethantšo sa Sesotho:
This table indicates that the same exercise was given to seven different groups of participants at different times due to their different locations. Consequently, Table 3.1 shows the results of all the participants from seven different groups at different locations. The number of participants in each location depended on the number of students available during the test. This means that the researcher used students that were given to her by the head of the department irrespective of the number that the researcher initially intended to utilise. For instance, in Mokhotlong there were 37 learners; Quthing 42; Qacha’s Nek 36; NUL 15; LCE 43; Mafeteng 43 and Leribe 38. This was done to avoid any inconveniences that the situation might have caused. For instance, if some students were given a test in the same class and others were left out, the teachers would have had a problem. The high schools that were visited excluded private schools, because all the government schools take Sesotho as a compulsory subject. The next table presents the results of students who utilised the Sesuto-English Dictionary.
The Sesuto-English Dictionary was used by different groups of students, thus, the number of students in this exercise was the same for both the dictionary and non-dictionary users. The use of dissimilar groups was done to ensure that the results were not influenced by the students’ experience gained while using the Sethantšo sa Sesotho. As the researcher was administering all of the tests, they were not given at the same time. The scores of Sesotho-English Dictionary users seem lower than those of the Sethantšo sa Sesotho users because the Sesotho English Dictionary does not have four of the words that appeared in the test.
The learners who were not using dictionaries managed to answer most of the questions compared to the dictionary users. The responses were disappointing, since most of the students who used the dictionary did not attempt to answer most questions, particularly learners from the LCE and Mafeteng. In the case of LCE, learners arrived late because they were writing a test just before the experiment. Similarly, the Mafeteng group also arrived late because they were writing tests before and after the experiment. It is assumed that they were not fully concentrating on the experiment. The table below presents the summary of the overall results for Question 2.
Unlike some Sethantšo sa Sesotho users who could not attempt all questions for Question 2, the Sesuto-English Dictionary users managed to answer all the questions under Question 2. As the use of a dictionary in this context has not been tested in any of the previous studies in learning a native language, this study intends to bridge this existing gap in the literature. Again, the decision to use experimental study was influenced by the fact that the previous studies, such as Hayati (2005), Laufer and Melamed (1994) and Atkins and Varantola (2008) to name a few, also used the same method to test the effectiveness of dictionary use by foreign language learners.
It was anticipated that the abovementioned method would assist the researcher to test the effectiveness of dictionaries by mother-tongue speakers, and that testing learners in this way is the best method of knowing how much learners know about their language. This method enabled the researcher to discover whether learners benefitted from using dictionaries while reading and writing or not. Interviews followed immediately after the tests.
Interviews are used as one of the methods of gathering data where particular groups or individuals are investigated. The researcher asks the subjects questions orally. The researcher can ask the participants questions that are related to the topic under investigation, such as people’s beliefs about the facts, their feelings, motives, present and past behaviours, standards for behaviour, and conscious behaviour for particular actions or feelings (Silverman, 1993). Interviews assist researchers to gather useful information from the respondents. The researchers may employ different types of interviews; this study chose to use focus groups. According to Rabiee (2004), a focus group is a method that involves the use of in-depth group interviews for which participants are chosen purposively. Participants focus on a given topic based on their knowledge of the topic under investigation to enable the researcher to elicit their opinion on the subject. Focus groups can provide diverse views, which are deeper and richer than the ones obtained from individual participants in a short time span (Thomas et al., 1995; Burrows & Kendall, 1997; Krueger & Casey, 2000). Thus, the interviews were useful for the purposes of this study as well as for time management.
It is assumed that the groups might provide valuable information because individuals would be sharing ideas during discussions and that might bring new information for the study. Babbie and Mouton (2001) mention that group discussions show direct evidence of the differences and similarities of participants’ experiences and opinions as compared to making such observations only when analysing statements derived from the interviews at a later stage. They also point out that the researcher is likely to miss some important information that individual members may have shared in face-to-face interviews. However, the researcher is of the opinion that since the interview is complementing the tests and the questionnaires, some information, which might be missed, would be covered by the information gathered from the tests and the questionnaires.
The selection of students in this study was purposive, as they are potential beneficiaries of the use of dictionaries. Dictionaries are mostly used in schools for various reasons, which include language learning, finding the correct spelling and meanings of words, translation, etc. The students’ contribution and concerns gleaned from the interviews would help lexicographers to see things differently, and that might force them to treat certain issues with great care when compiling dictionaries.
The students in each location were divided into two to four groups of 10 to 12 members each for the interviews. Learners who used the dictionary while reading and writing were grouped together and those who did not use the dictionary made their own group(s). The interviews were conducted immediately after the tests. The participants formed a circle and the 72 researcher moved around the circle to ensure that all the members of a group took part in the discussion. In cases where there were more than three groups, the researcher requested the teachers to help monitor the other groups. One member from each group was also asked to write down the answers to the questions. This decision was taken due to time restraints; the researcher was allowed 40 minutes for the interviews and 40 minutes for the test. It was therefore difficult for the researcher to monitor all groups simultaneously. The interviews that were not monitored by the researcher herself were tape recorded for purposes of accuracy of responses, and thereafter the responses were transcribed. This was used as an additional tool to complement their handwritten work.
Each group was asked to tell how they felt about the use of either the Sethantšo sa Sesotho or Sesuto-English Dictionary when learning Sesotho. Those who did not use a dictionary also shared their experiences based on what they felt during the test. However, Question 3 and Question 4 of the interview questions were relevant only for those who used the dictionaries. Non-dictionary users were provided with dictionaries to enable them to answer Question 5 (refer to Appendix 3 for these questions). The results helped the researcher to determine whether the dictionaries meet the needs of the current users, whether the learners benefitted from using Sesotho dictionaries during reading comprehension exercises and how they reacted to the availability of the dictionaries during the exercise.
About 23 groups were formed from 254 students who participated in each case. The number of participants varied per group ranging from 7 to 13 members in each group. When asked to tell how they felt about the use of Sesotho dictionaries when learning Sesotho, all 23 groups indicated that it was their first experience and that they were of the opinion that Sesotho dictionaries should be used during Sesotho lessons. Those who had access to the dictionary stated that they were able to learn words that were unknown to them and that their vocabulary was increased as a result. At that point, it was observed that the non-dictionary users were disadvantaged, but they were curious to know what the other group was learning.
Some of the reasons the respondents gave regarding dictionary usage when learning Sesotho were that it would improve their vocabulary acquisition and improve their proficiency level. They also mentioned that most of the words which occurred in their test were unknown, hence the need to use the Sesotho dictionaries in class.
On the question which required them to tell what they liked about each dictionary, the groups who had access to dictionaries mentioned that they were happy to realise that the dictionaries contained rich information which might help them know more about Sesotho words than when not using the dictionary. They were also of the opinion that words were explained in a clear and understandable way which made it easier for them to understand unknown words and that the dictionary clearly indicates the word class categories. In some cases, examples are provided on how the word could be used. Learners assumed that using dictionaries when writing Sesotho would contribute to the improvement of their writing skills.
Table of Contents
Table of Contents
CHAPTER ONE INTRODUCTION
1.1 Background to the study
1.2 Research problem statement
1.3 Aim and objectives
1.5 Significance of the study
1.6 Definition of key terms
1.7 Theoretical framework
1.8 Research design and methodology
1.9 Introduction of the two dictionaries under discussion
1.10 Ethical considerations
1.11 Chapter breakdown
CHAPTER TWO LITERATURE REVIEW
2.2 Types of dictionaries
2.3 Comparison of dictionaries
CHAPTER THREE RESEARCH METHODOLOGY AND DATA COLLECTION
3.2 Research population and sampling
3.3 Methods of data collection
CHAPTER FOUR A COMPARATIVE ANALYSIS OF SESUTO-ENGLISH DICTIONARY AND SETHANTŠO SA SESOTHO
4.3 The Sesuto-English Dictionary and the Sethantšo sa Sesotho compared
4.4 The strengths of the two dictionaries
CHAPTER FIVE A COMPARATIVE ANALYSIS OF SETHANTŠO SA SESOTHO AND SESUTO-ENGLISH DICTIONARY FROM THE PERSPECTIVE OF THE USERS
5.2 The effectiveness of the Sethantšo sa Sesotho and the Sesuto-English Dictionary in reading and writing
5.3 Significance of the Sesuto-English Dictionary and the Sethantšo sa Sesotho
CHAPTER SIX CONCLUSION
GET THE COMPLETE PROJECT