The research instrument is the standard term that researchers use for collecting data for enquiry. A distinction between two broad categories of instruments can be made, namely, researcher completed and respondent completed instrument measurement. An example of researcher completed instrument is the rating scale questions on survey questionnaires and a respondent completed instrument is the open ended survey questionnaire.
4.7.1 The procedures for conducting semi-structured interviews
Face-to-face, in-depth, semi-structured interviews were conducted with the chairpersons of the Councils, the representatives of the SRC’s on the councils and the Vice-Chancellors of the two educational institutions. Semi-structured interview schedules have, as with any other method of data collection, their strengths and weaknesses. The strengths are embedded in the following benefits as claimed by Van Teijlingen (2014). The first is the appropriateness when one wants to explore attitudes, values, beliefs and motives. Semi-structured interview schedules make it a very appropriate instrument for this study, because the values, beliefs, perceptions and attitudes of the internal population group can guide the development of a normative leadership model to guide organisational transformation at public institutions of higher learning. Secondly, a face-to-face interview gives a non-verbal indication which can enhance the trustworthiness of the response. Thirdly, a face-to- face interview can make sure that all questions will be answered and the participants can respond at their own pace. The researcher capitalised on these advantages throughout the semi-structured face-to-face interviews to reap the maximum benefit.
All methods, however, have disadvantages and the semi-structured face to face interview is no exception. The impact of the weaknesses was reduced or eliminated by doing proper prior preparation and by piloting the interview protocol with someone who was not part of the target sample. Such pilot testing is recommended, if possible (Saldana, 2011). The impact of social and cultural responses can never be disregarded and to reduce this impact questions were constructed prior to the interviews. The privacy of the participant can be affected with face-to-face interviews and to reduce the impact of this weakness the researcher invited the participants to suggest a venue, time and date, within a certain time frame of two weeks, most comfortable and convenient for them.
Owing to distance, one of the interviews was conducted via video conferencing while five of the semi-structured interviews were conducted face-to-face in the offices of the participants. The participants may have prejudices, stereotyping, and perception of the researcher which may impact their response (Van Teijlingen, 2014). To address possible prejudices the researcher was well prepared, objective and maintained a rapport with the participants. When difficulties in understanding the meaning of the response were encountered this was addressed through probing and short questions. This applies also when clichés or metaphors or analogies used by the participants are not generally accepted language, and the researcher requested the participants to explain what they meant. This was done to make sure that the data obtained are a true reflection of the participants’ responses.
Prior to the conducting of the in-depth, face-to-face semi-structured interviews the researcher acknowledged their disadvantages to reduce or/ and to eliminate this he took the following precautions as recommended by Silverman (1993), Shank (2006) and Creswell (2014). Prior to the interview, the researcher established relationships with the Vice-Chancellors, chairpersons of councils and the representatives of the SRC’s on the councils. This was done via informal meetings with the people to be interviewed or via their gatekeepers or personal assistance. This strategy was chosen, because of the concerns raised by Kvale and Brinkman (2009) who address the issue and importance of access to the elite. In this study the entire sample groups are classified as elite. The construction of the questions for the interview and knowledge of the literature from the side of the researcher were of crucial importance. Warm-up questions and interview questions were prepared prior to the interviews. This was done to eliminate the possibility that different questions would be asked to different participants, which will increase the risk to do cross participant comparisons. The second reason was to make certain that the interview would move in the desired direction. Cultural differences and professional positions of the participants were considered when preparing the questions and during the interview. Differences can be caused by culture, age and professional positions. This is of utmost importance as different cultural groups perceive issues differently which has an impact on their response.
To ensure that no distraction or interruptions occurred the researcher requested the participants to suggest a venue, time and date, during the month of July to September 2017, to ensure the comfort of the participants. When the sample was selected the researcher tried to make sure that the sample will disclose the information the researcher was seeking. This was achieved by sampling executive management. The researcher confirmed the appointments for interviews at least five working days in advance and followed it up with a reminder on the day of the interview. The interviewees were told in the informed consent request letter that the interviews would be audio recorded and that the transcribed interviews and audio recordings would be destroyed five years after the dissertation was approved and placed in the public domain. To eliminate the malfunction of technology (Saldana, 2011) the researcher recorded the interviews with an audio tape recorder and on his private mobile phone. After the interviews were transcribed and verified for verbatim transcribing, the researcher placed the tape recorder in a safe place and deleted the recordings from his private mobile phone.
During the interview the researcher was respectful, maintained a rapport with the interviewee and showed an interest in what the interviewee was saying. The researcher also respected the status of the interviewees. Interviewees were allowed to choose their own way of expressing their opinions, thoughts and ideas. Throughout the interview the researcher continuously reminded himself that although the interviewee may be very convincing the responses may only be opinions or perceptions and not facts. Metaphors or clichés used by the respondents during the interview were never interpreted by researcher, but the researcher rather asked for an explanation.
4.7.2 The procedures for administration of the survey questionnaire
The purpose of correct procedure for the administration of the questionnaire was to reduce personal bias in the responses of the Vice-Chancellors, chairpersons of councils and SRC representatives on council, which appeared to be a challenge emerging from self-perception assessments. This self-perception assessment correspond with Ellaad (2003) and Walfish, Mc Allister, O’Donell & Lambert (2012). The strength of this method of data collection lies in the fact that it is the easiest and least time consuming method for the researcher to gain qualitative data (Debois, 2016). It is possible to get responses which are both open and closed. Survey questionnaires are in written format which means the researcher does not need to confirm their correctness with the respondents. Another advantage is that a large number of participants can be involved in a short time frame. The analyses of data from survey questionnaires are relatively easy and can be done within a limited time
The weaknesses of a survey questionnaires according to Debois (2016) are that misinterpretation of questions may result in different responses. To eliminate this possibility, proper care was taken with the development of the questions for the survey questionnaire. The richness of the response may be affected by the fact that a survey questionnaire cannot convey any feelings of the participants. Some participants may have hidden agendas when completing the survey questionnaire. This may negatively reflect their responses that ultimately affect the reliability of the data. Respondents may skip answering some questions on the survey questionnaire. The reasons can be the following: time available; fear for victimisation; respondents do not properly understand the question(s), or respondents do not have an opinion/response to the question. This was eliminated by constructing questions that eliminated any bias and by giving participants a reasonable time to complete the survey questionnaire.
The following points, recommended by Leady and Ormrod (2015), were observed when constructing the survey questionnaires. The content of the survey questionnaire was short, concrete and unambiguous to avoid response bias. The respondents needed time to complete the survey questionnaires and as time is precious the respondent did the researcher a favour to complete it. The participants for this survey were very highly professional people with a very busy work schedule. The second important issue that the researcher considered was whether he actually needed the information to solve the research problem. To avoid this concern, questions that can provide answers to the research problem were asked. Black (1999) suggests that one should randomly distribute half positive and half negative questions in a survey questionnaire. This will lessen the temptation for the respondent to just mark all of them the same, which will reduce bias.
The researcher took all actions possible to avoid leading questions, to ensure the consistency of questions for the survey questionnaire and that the survey questionnaire looked professional. The participants of the survey questionnaires were educated professional people and may read between the lines. In extreme cases the participant could withdraw from the process of data collection and was aware of this option. The coding for responses was validated during the setting of the questions for the survey questionnaire. To ensure the validity of the survey questionnaires two actions were taken. The researcher scrutinised the questions by putting himself in the place of the participants and secondly two pilot tests were done. The purpose was to determine if the survey questionnaire was clear and that the preferred responses could be expected.
Black (1999) distinguishes between the binary and five scale rating scales and open ended questions. In the section on employment information this study opted for binary questions. The survey questionnaire chose a Likert five scale point questions with options for the respondents to access statements for Section B. Options ranging from strongly agree, agree, neither agree nor disagree, disagree and strongly disagree. Free-response or open ended questions were asked to validate opinions on the five Likert scale e question.
The hard copy semi-closed ended survey questionnaires were delivered in sealed envelopes in person by the researcher to the academic deans, the Deputy Vice-Chancellor: Academic Affairs and Research (NUST) and the Pro Vice-Chancellor: Innovation and Research and Development and the Pro Vice-Chancellor: Academic Affairs (UNAM). To ensure confidentiality of the responses to the survey questionnaires an envelope for return was put in a sealed envelope containing the survey questionnaires. A pre-designed register was used for the respondents or gatekeeper to sign for receiving and returning of the survey questionnaires. After the survey questionnaire was completed, the respondent put it in the enclosed envelope and sealed it for collection. The researcher collected the sealed envelope in person from the respondents or their gatekeepers.
The return rate for a mailed survey questionnaires is 50% or less while the e-mailed survey questionnaires rate is even lower (Rogelberg & Luong, 1998 and Sheehan, 2001). The researcher chose to deliver the survey questionnaires in person to the Pro Vice-Chancellor and Deans. Ten working days was mutually agreed upon for the completion of the survey questionnaires. The time was also arranged with the respondent or their gatekeeper for the collection of the completed survey questionnaires. The following strategies were also put in practice to increase the rate of return. Telephone calls were made and emails sent to respondents to make certain that the survey questionnaires were completed on time. The timing for the administration of the survey questionnaires is important to increase the return rate. A week after the start of the second semester the respondents do not have such a busy working schedule as at the end of a semester. Prior to the delivering of the survey questionnaires an appointment was made to explain the purpose of the survey questionnaires and a word of motivation was given to complete these survey questionnaires. The importance of truthfulness and answering all questions in the completion of the survey questionnaire was also emphasised. The ethical issues that would be observed with the administration of the survey questionnaires and data were outlined in the informed consent letter to respondents.
The letter of informed consent together with the Ethical Clearance Certificate and letter of consent (NUST respondents) were mailed to all the HODs of UNAM and NUST. The survey questionnaires were then sent via SurveyMonkey to all HODs. Phone calls were made to ensure that all the respondents had received the survey questionnaires via SurveyMonkey and to explain the importance of participation. To improve the electronic administration of the survey questionnaires the researcher employed the function of SurveyMonkey to regularly remind respondents to complete the survey questionnaires. Phone calls were made and where possible personal visits were conducted to counteract the usual low response rate of electronic survey questionnaires.
For the secondary data the annual reports and institutional reports from 2012 to 2016 of NUST and UNAM were studied. These reports were studied in conjunction with the acts and the statutes that govern these two public institutions. The aim of the secondary data was to complement the richness of the primary data and to determine if any violation of the acts of UNAM and NUST had occurred.