CHAPTER THREE METHODOLOGY
A survey relating to assessing nursery/garden centre compliance with regard to invasive alien legislation has never been carried out in South Africa. As a result, no secondary data pertaining to local nurseries/garden centres and the compliance with CARA section 15 and the proposed NEMBA regulations, section 70 are available.
Therefore, primary data will be required to address the research problem specified in Chapter 1. This chapter will discuss the proposed methodology for this study.
Creswell (2009, p.175) notes that quantitative research is “a means for testing objective theories by examining the relationship among variables. These variables can be measured on instruments, so that numbered data can be analysed using statistical procedures”. He further explains that, when using the quantitative approach, a researcher will “concentrate on the quantitative facts or data associated with the problem and develop mathematical expressions that describe the objectives, constraints and other relationships”.
“Qualitative research refers to the meaning, definition, analogy, model or metaphor characterising an occurrence” (Creswell, 2009, p.180). Cassel and Symon (1994, p.125) suggest that qualitative research focuses on the following aspects:
Interpretation rather than quantification;
An orientation towards process rather than outcome;
A concern with context regarding behaviour and situations in forming experiences; and
An explicit recognition of the impact of the research process on research situation. This study followed a mixed methods approach as the study required both the quantitative analysis of questionnaires and the interpretation of interviews. The survey was a means to gather statistically valid quantitative data and the interviews were aimed to collect in-depth qualitative data in terms of perception, idea or explanation through participant‟s expressions (Creswell, 2009, p.190).
Mixed Methods Research
Many terms are used for this approach, such as the following: integrating, synthesis, qualitative and quantitative methods, multimethod, mixed methodology and mixed methods (Creswell, 2009, p.152). According to Creswell (2009, p.153), mixed methods focuses on combining both quantitative and qualitative research methods in a research study. It involves philosophical assumptions and the use of quantitative approaches so that the overall strength of a study is greater than either quantitative or qualitative alone.
This approach has many applications in diverse fields in social and human sciences, and is emphasized in journals such as the Journal of Mixed Methods Research, Journal of Social Research and Qualitative Health Research (Creswell, 2009, p.150). Creswell (2009, p.150) explains that researchers employ a mixed methods design to broaden understanding by incorporating both quantitative and qualitative methods or use one approach to better understand, explain, or build on the results from another approach. Further “mixing” of quantitative and qualitative approaches might be within one study or among several studies in one programme of inquiry. “Mixing” of the two types of data might occur at several stages within a study, i.e.; the data collection, the data analysis, the interpretation, or at all three phases (Creswell, 2009, p.150).
Creswell (2009, p.152) further states that, when planning a mixed methods study, four aspects to consider are; timing, weighting, mixing, and theorising.
Timing in mixed methods research can be sequential, where quantitative and qualitative data are collected in phases or concurrent, where both types of data are collected at the same time or phase in the study. Weighting is the priority given to quantitative or qualitative research in a particular study. In some studies, the weight might be equal; in others it might emphasise quantitative or qualitative data. Mixing means either that the quantitative or qualitative data are, intergraded at the beginning of the study, or kept separate, and combined at the end of the study or combined in some way during the study (Creswell, 2009, p.152).
Creswell (2009, p.154) also identified the following means of mixing data in mixed methods research: Connecting: involves connecting the data from one phase of the study to another. Quantitative and qualitative are connected between a data analysis of the first phase of research and collection of the second phase; Integrating: both sets of data are merged through a comparison approach or through data transformation; and Embedding: a secondary form of data is lodged within a larger study with a different form of data as the primary database. The secondary database provides a supporting role”.
Theorizing in mixed methods research can be described as an advocacy perspective that shapes the types of question asked, informs how data is collected and analysed, and provides a call for action or change. Theory may include deductive quantitative testing and verification or inductive qualitative theory (Creswell, 2009, p.158).
Creswell (2009, p.159) further describes six strategies in mixed methods research that is organised around how data is collected, viz.: Sequentially (Sequential explanatory strategy and Sequential exploratory strategy); Concurrently (Concurrent triangulation strategy and Concurrent embedded strategy); Transformative lens (Concurrent transformative strategy and Sequential transformative strategy).
Application of Mixed Methods in this Study
The aim of this study was to assess the local horticultural industries‟ compliance with CARA section 29(15), and the proposed NEMBA legislation section 70. This researcher addressed the extent to which the local horticultural industry was playing their role in combating the spread of IAPs, as well as issues regarding communication between relevant Government agencies and the horticultural industry and to the extent that local Industry has been included in governmental initiatives to address the IAPs problem.
Following a “Concurrent Embedded Strategy”, a questionnaire was administered to sales personnel of nurseries and garden centres, within the selected study area and interviews were conducted with nursery and garden centre owners and senior managers to gauge their feelings and perceptions with regards to the problem discussed in Chapter 1. Tashakkori and Teddlie (2003, p.517) describe this approach where employees could be studied quantitatively and managers could be interviewed qualitatively, as a multilevel design. This researcher made appointments with the respondents and conducted these interviews in English. Observations were carried out with the aid of an observation checklist, about what nurseries/garden centres are selling during on-site visits to a sub-sample of nurseries/garden centres within the selected study area.
Quantitative and qualitative data were gathered at the same time or phase of the study and were analysed simultaneously. “Collecting and analysis of both sets of data can be a rigorous and timeconsuming process” (Creswell, 2009, p.206). The “concurrent embedded model” emphasises a major primary form of data collection (e.g. surveys), and includes a minor secondary form of data collection (interviews) (Creswell, 2009, p.206). He explains that the fact that both forms of data collection are not equal in size and rigour enables the study to be reduced in scope and manageable for the time and resources available.
Figure 3.1 shows that the “weighting” of this mixed methods study were assigned to quantitative data collection and analysis, and the notation for the study were “QUAN + qual.” A “concurrent embedded” approach has a primary method (in this study quantitative, QUAN) that guides the project and a secondary method (qualitative, qual) that provides a supporting role in the procedures (Creswell, 2009, p.207).
The “mixing” of the two data sources occurred during the discussion of survey and interview results, at the interpretation stage of the research process. This researcher adopted a comparative approach and compared the quantitative results and analysis of the questionnaire and observations with the findings of the interviews.
Capitalisation indicates the weight or priority of the quantitative data, analysis and interpretation of this study;
“QUAN” and “qual” stand for quantitative and qualitative respectively. The same number of letters were used to indicate equality between the forms of data;
A “+” indicates a concurrent form of data collection, with both quantitative and qualitative data collected at the same time; and
A “QUAN” “qual” notation indicates that qualitative methods are embedded within a quantitative design as shown above.
Area of Study
The area of study comprised of selected nurseries and garden centres within EThekwini (Durban) and uMsunduzi (Pietermaritzburg) geographical regions of KwaZulu-Natal. Through listings in the Yellow Pages telephone directory, the South African Gardening magazines and the South African Nursery Association membership list, 75 nurseries/garden centres were originally identified in this study area.
Since the above-mentioned listings were posted, 26 nurseries/garden centres had closed down. An additional 11 nurseries/garden centres were identified through “snowball” sampling (www.statpac.com/surveys/sampling.htm – 8k. circa, 2009), during the data collection process. A total 60 nurseries/garden centres participated in this mixed methods study.
In order to arrive at vital and reliable conclusions, adequate sampling of the population in the study area needs to be done (Tashokkori and Teddlie, 2003, p.715; Mugo, circa, 1995). But if a population is “so small” that sampling parts of it will not provide accurate estimates of a whole, a census is the only way to get accurate information (Salant and Dillman, 1994, p.6). It is also important to take response rate and non-response bias into account (Draugalis and Plaza, 2008, p.142). It becomes important to get a high response rate in small populations (Krejcie and Morgan, 1970, cited in Draugalis & Plaza, 2008, p.144). Table 3.2 characterises this requirement.
Seventy five nurseries and garden centres have been collectively identified in the EThekwini (Durban) and uMsunduzi (Pietermaritzburg) study area, and for the purpose of this mixed methods study, one response represented one nursery/garden centre. A census (approach all the nurseries and garden centres within the target area) was conducted for the quantitative strand of this study, taking the response rate and non-response bias into consideration. Due to factors mentioned in section 3.5, 60 nurseries/garden centres participated in this study. Procedures followed to “Gain Entry to Research Sites” (refer 3.7) ensured a positive response rate.
A purposive sampling method was selected for the qualitative strand of this study. “Purposive sampling is a form of non-probability sampling, where units (individuals, institutions) are selected based on specific purposes associated with answering a study‟s research questions” (Teddlie and Tashakkori, 2009, p.343).
Non-probability sampling does not provide any basis for estimating the probability of items in the population for being included, sampling error cannot be measured and researcher bias has a great chance to enter (Sridhar, 2008). However, non-probability sampling is suitable for small in-depth inquiries, where sampling permits the selection of interviewees whose qualities or experiences express an understanding of the phenomena in question (Sridhar, 2008). It is for this reason that this researcher conducted interviews with selected senior managers/owners of nurseries and garden centres as well as with stakeholders in pertinent government agencies and trade associations, to address the research questions in chapter 1.
In most cases, quantitative studies require mathematically defined procedures that allow one to estimate the characteristics of the population within a prescribed margin of error, usually +/- 5% or 1% (Teddlie & Tashakkori, 2009, p182). This kind of precision is not possible with most qualitative studies, as the “sample sizes used in qualitative research are typically so small that they are transferable to only a small sampling frame” (Teddlie & Tashakkori, 2009, p.182). According to Patton, (2002, p.244), “There are no rules for sample size in qualitative inquiry.”
Mixed methodologists have combined probability and purposive sampling techniques, to meet the specifications of mixed methods design. For example; “Stratified purposive sampling” and “Purposive random sampling”, are types of purposive sampling techniques, but they also include a component of probability sampling, viz., stratified and random (Teddlie & Tashakkori, 2009, p.182).
Purposive random sampling involves taking a random sample of a small number of units from a larger target population (Kemper, Stringfields and Teddlie, 2003, p.375). This method of sampling adds credibility to the evaluation by generating qualitative process oriented results to complement the large scale, quantitative oriented research (Teddlie &Tashakkori, 2009, p.186).
Purposive random sampling selection for the qualitative strand of this current mixed methods study proceeded as follows:
Step 1: The target population was divided into two categories, viz., nurseries and garden centres (Refer to 3.8 Differences between Nurseries and Garden Centres);
Step 2: 20 Nurseries and garden centres were then randomly selected, using the random numbers table (Siegel &Castellan, 1988, p.322); and
Step 3: Nurseries‟ and garden centres‟ categories were then divided into four sub-categories, viz.; Trade Affiliated Nurseries, Non-Trade Affiliated Nurseries, Trade Affiliated Garden Centres and Non-Trade Affiliated Garden Centres.
As there were no clearly established standards for how large the sample should be to generate trustworthy results (Teddlie & Tashakkori, 2009, p.185), the sample size of this qualitative strand, interviews were done until saturation was reached.
In stratified purposive sampling, the stratified strand is similar to probability sampling and the small number of cases it generates is characteristic of purposive sampling (Teddlie & Tashakkori, 2009, p.186). Due to the sensitive nature of this enquiry [nurseries and garden centres do not want to be labeled as selling IAPs and therefore not compliant with IAPs legislation], this researcher anticipated resistance from some owners/managers. This researcher did not want to create the impression of wanting to police any nursery/garden centre. Therefore, observations were merely conducted to determine what nurseries/garden centres stock and sell, using the stratified purposive sampling technique.
Gaining Entry to Research Sites
“One of the many problems facing researchers aiming at doing research studies into organisations is gaining access”, even more so if the topic being researched focuses on a sensitive matter (Okumus, Altinay and Roper, 2007, p.9) As mentioned before, the topic of this mixed methods study is sensitive in that nurseries/garden centres do not want to be seen as non-compliant with AIPs legislation and thus contravening the law.
It is also the perception amongst nursery men that they have been excluded in the whole process around issues regarding governmental programmes and initiatives pertaining to AIPs eradication and awareness (Peter Rowels, personal communication, September16, 2009).
According to Peter Rowels (Personal communication, September16, 2009), Industry was only informed about CARA legislation about three weeks before the regulations were passed by Government. The law required plant species that had been legislated against, in terms of CARA regulations, to be removed from the sales floor and destroyed. The Industry incurred substantial losses and Government did not offer any compensation for these losses (Peter Rowels, personal communication, September16, 2009). Also, there seems to be much confusion among nursery men regarding which plants are “allowed” and which not (Jo-Anne Hilliar, personal communication, September 21, 2009). For example: Lantana camara hort. Common names; Lantana, Tickberry or Cherry pie, is a declared category 1 (CARA) alien species (www.sana.co.za/Alien -Invasive-Plants/invasive-alien -plants-cara-list, circa, 2009). It is the perception of Industry (Hester McLachlan- Evans, personal communication, February 24, 2010) that the creeping, yellow Lantana cultivar, Lantana montevidensis „Sundancer‟ does not set seed and is ,therefore, regarded as “sterile” and non-invasive (Southern African Plant Invader Atlas, 2009). Many „Sundancers‟ are sold in nurseries/garden centres across the country. However, studies in South Africa and Australia found that „Sundancer‟ is part Lantana camara hort. and is only “sterile” if planted in isolation, as it can cross pollinate with L. camara, and occasionally produce seed (Southern African Plant Invader Atlas, 2009).
Now, in terms of CARA, “All seed producing species or hybrids of Lantana that are non-indigenous to Africa,” are classed as category 1 invaders and are illegal to grow, propagate or sell in South Africa (www.sana.co.za/Alien -Invasive-Plants/invasive-alien -plants-cara-list, circa, 2009). However, there is no evidence that „Sundancer‟ is invasive (Southern African Plant Invader Atlas, 2009), but if „Sundancer‟ occasionally produce seed, is it illegal in terms of CARA? According to the South Africa Government Gazette (No. 32090, April 3, 2009), “All seed producing species or seed producing hybrids of Lantana that are non-indigenous to Africa” are legislated against.
Much controversy exists around some plants that are on the proposed NEMBA, Table X list (Hester McLachlan- Evans, personal communication, February 24, 2010). For example; “Murraya paniculata (L)”, also known as “Murraya exotica L., common name; Oranjejasmyn or Mock Orange, a category 1b invader in KwaZulu–Natal, Mpumalanga, Limpopo and Eastern Cape” (South Africa Government Gazette No. 32090, April 3, 2009). In terms of NEMBA category 1b species require “compulsory” control must be removed and destroyed and no permits will be issued for these species to be grown, propagated or sold in South Africa (South Africa Government Gazette No. 32090, April 3, 2009). Industry feels that Murraya paniculata (L) should not be legislated against; as there is not enough evidence to prove that the plant is a potential invader (Hester McLachlan-Evans, personal communication, February 24, 2010). Jacque Malan, SANA Biodiversity Representative, under the direction of Industry, has logged an appeal against Murraya paniculata (L), being on Table X. Industry is awaiting the outcomes (Hester McLachlan-Evans, personal communication, February 24, 2010). Many plants species on the proposed Table X list are popular ornamental plants. When NEMBA regulations are enacted, it might have a substantial financial impact on Industry.
In view of the above, nurseries/garden centres might perceive this current study with suspicion as to the intentions of the research. Also, this researcher did not want to give the impression as to want to police anyone. Therefore, it was imperative to the study that this researcher gains the trust of the participants.
In the opinion of Okumus, et al. (2007, p.10), to gain the trust of participants, a researcher must first gain the trust of the “gatekeepers.” Gatekeepers are people of authority, like business owners, managers and individuals who can provide and facilitate access for the researcher. These authors also noted that gatekeepers may deny access because researchers might fail to provide answers about what, how and why they will carry out a specific study and whether the study will be of any value to managers themselves and also the company.
Okumus et al. (2007, p.10) also recommend that to gain access, the researcher and the gatekeepers must come to an agreement in terms of what, when and how data are collected and what might be returned. The researcher must also develop a good understanding and collaboration with the relevant managers.
This researcher gained access to the selected research sites, by implementing the following procedures: First, the target population was grouped according to their geographical location within the area of study; Then the researcher made telephonic contact with owners/managers, introduced herself and gave a brief background and purpose of the study; After permission to conduct the enquiry was granted, a more comprehensive background of the study together with an Informed consent letter was sent via email and faxed to those that did not have an email address; and A day before the researcher planned to visit nurseries/garden centres in a specific geographical location; the researcher telephoned those nurseries/garden centres and informed owners/managers about the planned visit.
Difference between Nurseries and Garden Centres
The horticultural industry is a diverse industry, displaying both retail and production functions with a variety of products ranging from services to goods (Enright & McDonald, 1997), and can be divided into wholesale and retail sectors. Wholesale nurseries, propagate, produce and maintain plants for sale to retail nurseries, supermarkets, garden centres and landscapers. Retail nurseries market and promote plants, products and services, as well as maintaining plants for sale to the general public (Sellmer & Dana, circa, 1993).
Traditionally, retail nurseries‟ core business was to sell plants and gardening-related products such as fertilizer, compost, potting mixtures and garden tools (Regan, circa, 2002). According to Regan, (circa, 2002) many nurseries may buy in smaller plants from specialised growers and then raise them into saleable plants; often nurseries would also grow and propagate their own plants on a small scale. Regan (circa, 2002) further stated that during the late 1990‟s, nurseries were prospering, benefiting from a strong construction market, rising household incomes and growing interest in landscape aesthetics and environmental enrichment. This gave rise to a new generation of retail outlets; the garden centre. Like nurseries, the core business of garden centres is the selling of plants and garden related product (http://gardencenternursery.wsu.edu/site/RetailNurserySiteSelection.html, circa, 2002). The main difference was the emphasis on the supply of “lifestyle products and services” such as outdoor furnisher, gift lines and landscaping services. Some garden centres even have coffee shops, petting zoos and children‟s play area‟s to add value to their establishments, offering clients a one-stop shop to spend a leisurely time shopping (http://gardencenternursery.wsu.edu/site/RetailNursery SiteSelection.html, circa, 2002).
Conversely, the garden centre developed a whole different image to a nursery, with the emphasis on the design of the structures and buildings, surrounded by lawns, paving and well-kept display gardens (http://gardencenternursery.wsu.edu/site/RetailNurserySiteSelection.html, circa, 2002), better plant signage, clearly marked parking bays, and a more diverse product range. Another marked difference between nurseries and garden centres is that garden centres do not grow and raise plants on site, but buy in saleable plants from wholesalers and specialised growers (Regan, circa, 2002).
For the purpose of this study, nurseries and garden centres were differentiated as follows: Nurseries: Core business – selling plants and garden-related products, with no added value lifestyle products and services affiliated to the business. Some nurseries may buy in smaller plants and raise them into saleable plants or grow and propagate plants on a small scale on site; and Garden Centres: Core Business – selling plants and garden related products, with added value lifestyle products and services affiliated to the business. Garden centres do not raise or grow plants on site.
Quantitative Data Collection
Quantitative data for this study were generated through a questionnaire and structured on-site observations. Data collection tools were in English.
The survey was limited to sales personnel of nurseries and garden centres. Sales personnel can include managers and owners depending on the individual business structure. The survey populations were assumed to be people who had a good grasp of the English language, as most horticultural literature and plant reference material are only available in English. Sales personnel are also well positioned to address customer queries, recommend appropriate plant choices and should have an extensive knowledge as to what plants their individual business stocks and sells.The questionnaire was a self–administered data collection tool specifically designed for this study. According to Dillman (1978), self–administered questionnaires offer a truer reflection of points surveyed; the responses are, therefore, objective. Dillman (1978) further explains that, “respondents to self–administered questionnaires are relatively unlikely to answer to please you because they believe certain responses are more socially desirable”.
Prior to delivering the questionnaire, this researcher made contact with managers and/or owners (gatekeepers) of the selected sample of nurseries and garden centres to gain permission to deliver the questionnaire and to allow sales staff to complete it the same day.
Once permission to gain entry to research sites (refer to 3.7) was granted, a comprehensive background of the study, together with an Informed consent letter were send via email and faxed to those that did not have an email address. “When survey questionnaires are used in a study, the researcher is employing a strategy in which participants use self-report to express attitudes, beliefs, and feelings towards a topic of interest” (Teddlie & Tashakkori, 2009, p.232). The questionnaire for this mixed methods study was a closed– ended quantitative questionnaire. Teddlie and Tashakkori (2009, p.232) explains, that “items with closed–ended responses are more efficient to collect and analyze,” than items with open–ended responses. Response formats associated with closed–ended questionnaires include Likert scales, checklist and rank orders (Teddlie & Tashakkori, 2009, p.232).
TABLE OF CONTENTS
LIST OF APPENDIXES
LIST OF TABLES
LIST OF FIGURES
CHAPTER ONE SCOPE OF THE STUDY
1.2 The Research Problem
1.3 The Aim of this Study
1.4 Research Objectives, Questions and Hypotheses
1.5 Chapter Outline
CHAPTER TWO LITERATURE REVIEW
2.2 Impacts of Invasive Alien Plants
2.3 Invasive Alien Plant Legislation in South Africa
2.4 Voluntary Codes of Conduct for the Horticultural Industry
2.5 Key Stakeholders in the Control of Invasive Alien Plants
2.6 Management Challenges in the control of IAPs
2.7 In Summary
2.8 Gap in Literature
CHAPTER THREE METHODOLOGY
3.2 Research Design
3.3 Mixed Methods Research
3.4 Application of Mixed Methods in this Study
3.5 Area of Study
3.6 Sampling Techniques
3.7 Gaining Entry to Research Sites
3.8 Difference between Nurseries and Garden Centres
3.9 Quantitative Data Collection
3.10 Qualitative Data Collection
3.11 Validity and Reliability
3.12 Ethical Considerations
CHAPTER FOUR DATA ANALYSES
4.2 Quantitative Procedures
4.3 Qualitative Data Analysis
CHAPTER FIVE RESULTS AND DISCUSSIONS
5.2 Descriptive Statistical Results
5.3 Inferential Statistical Results
5.5 Thematic Analytical Results
5.6 Integration or Mixing of Results
5.7 Concluding Remarks
CHAPTER SIX CONCLUSIONS AND RECOMMENDATIONS
GET THE COMPLETE PROJECT
AN INVESTIGATION INTO THE COMPLIANCE OF SELECTED NURSERIES AND GARDEN CENTRES WITHIN KWAZULU-NATAL ETHEKWINI AND THE UMSUNDUZI GEOGRAPHICAL REGIONS, WITH THE CONSERVATION OF AGRICULTURAL RESOURCES ACT 1983 (ACT NO. 43 OF 1983) CARA AND THE NATIONAL ENVIRONMENTAL MANAGEMENT:BIODIVERSITY ACT 2004 (ACT NO. 10 OF 2004) NEMBA.