Questionnaire survey with students

A questionnaire survey was chosen, as this method can provide a rich source of varied material (Welman, Kruger & Mitchell, 2005:175). As this method of data collection ensures participants’ anonymity, the questions are more likely to be answered honestly. Questionnaires remove the issues related to research bias, as there is no personal contact and the participants cannot be manipulated in any way (Neuman, 1997:239).

Questionnaires aimed at the PAS 111 group of students were designed taking Louw’s (2005) framework into consideration. Louw (2005) focused on academic factors that might potentially influence student departure from agricultural training institutes. For this study, closed- and open-ended questions were used. A closed-ended question is one where participants in the research are confronted with a question, but instead of being able to give any answer, they are provided with fixed response options from which to choose (Neuman, 1997:240). An open- ended question is one where the researcher asks the participant a question to which any answer can be provided.

91

Table 4.3 gives a summary of the advantages and disadvantages of using closed-ended questions, which were taken into account during the construction of the questionnaire for this study.

Table 4.3: Advantages and disadvantages of using closed-ended questions (adapted from Neuman, 1997:241)

Advantages Disadvantages

Easier and quicker for participants to answer

Research can suggest ideas participants would not otherwise have

Answers of different participants are easier to compare

Participants with no opinion or knowledge can still answer

Answers are easier to code and statistically analyse

Participants may become frustrated when their desired answer is not a choice The response choices can clarify question

meaning for participants

Confusing if too many response choices are provided

Participants are more likely to answer about sensitive topics

Misinterpretation of a question can go unnoticed

Fewer irrelevant or confused answers Distinction between participant answers may be blurred

Less articulate or less literate participants are not at a disadvantage

Clerical mistakes or marking the wrong response is possible

Replication is easier Participants are forced to give simplistic responses to complex issues

Table 4.4 provides a summary of the advantages and disadvantages of using open-ended questions, which were taken into account during the construction of the questionnaire and interview questions for this study.

92

Table 4.4: Advantages and disadvantages of using open-ended questions (adapted from Neuman, 1997:241)

Advantages Disadvantages

Unlimited number of possible answers Different participants give different degrees of detail

Participants can answer in detail and clarify responses

Responses may be irrelevant or buried in useless detail

Unanticipated finding can be discovered Comparisons and statistical analysis become difficult

Adequate answers to complex issues are provided

Coding of responses is difficult

Creativity, self-expression and richness are permitted

Articulate and highly literate participants have an advantage

Participants’ logic, thinking process and frame of reference are revealed

Responses are written verbatim, which is difficult for the researcher to interpret More time, thought and effort are necessary Participants may be intimidated by

questions

Table 4.4 illustrates how complex the undertaking of interviews may be when using open- ended questions. Open-ended questions were opted for in this study with careful consideration of the disadvantages and ethical issues (see Section 4.5) to enhance data accuracy and validity.

Tables 4.3 and 4.4 above demonstrate Plowright’s (2011) observation that any particular method has various advantages and disadvantages. Neuman (1997:240) explains that it is not a matter of which method is best, but rather which method is best suited for the conditions of the research and addressing the research question. Despite the disadvantages of using questionnaires, the advantages outweighed the disadvantages and a questionnaire was found to be a valid method of data collection for this study. Many researchers suggest that anyone who seeks to use a method should employ a combination of the above techniques in order to strengthen the interpretations and findings with regard to the reliability and validity of the research (Babbie & Mouton, 2001; Creswell, 2007; Flick, 2009; Silverman, 2001). Henning

93

(2004:147) sums it up aptly: “I would also like to promote good craftsmanship, honest communication and action as reasons for rating research as good scholarship.” I have attempted to do this in this study. The instruments used are discussed next.

a. Data generation

The participants in this study were drawn from the 82 registered first-year BAgric students in 2013. A total of 82 students were registered for the PAS 111 module of 2013. The questionnaires were distributed to all students in this group while they were assembled in one venue on the EATI campus during a normal calendar day in August 2014. As explained in the previous section, the questionnaire contained closed- and open-ended questions (cf. Addendum A). Before distributing the questionnaires, I explained the purpose of the research and provided a brief overview of the purpose of the research. This was done to ensure that all students had a uniform understanding of the research topic. The students were also informed that their participation is voluntary and anonymous. There were no clarification questions asked by students and they could complete the questionnaires immediately, which took approximately 30 minutes on average. A total of 52 questionnaires were returned. There were 30 students who chose not to participate. Refer to Addendum D for an example of a completed questionnaire.

b. Data analysis

Once the questionnaires had been collected I organised the raw open-ended narrative data into conceptual categories in order to identify emerging themes, which were used to analyse the data, as recommended by Neuman (1997:421). As themes emerged, they were coded and plotted using an Excel spreadsheet. Open coding was used for this study, which entails locating themes and assigning initial codes to each theme in an attempt to make the data manageable. The identified themes are assigned codes, but the possibility still exists for creating new themes and changing existing codes. The purpose of the coding is to make the amount of data manageable and meaningful (Welman et al., 2005:213).

The areas that were investigated were the students’ demographic information, educational background, academic performance (closed-ended questions) and academic performance (open-ended questions). Topics under the closed-ended section of academic performance that were investigated were admission requirements, language, interest, class attendance, study time, contact time with the lecturer, fairness of the examination paper, whether the learning

94

material correlated with the presentations, level of presentation, student support and benefit of tutorials and summer schools. Topics under the open-ended section of academic performance were challenges in Chemistry, factors that helped the student succeed in Chemistry, suggestions on what can be done to succeed in Chemistry, how to overcome challenges in Chemistry, reasons why students fail in Chemistry and approaches to learning Chemistry.

During analysis of the questionnaire data, no attempt was made to generalise the data, as the population and sample were small and this study only focussed on one institution, therefore generalisation to all agricultural higher education institutions based on this study would not be valid.