Tshikombeni knowledge of national biodiversity symbols in South Africa

Description

Other

Funding provided by: National Research Foundation
Crossref Funder Registry ID: https://ror.org/05s0g1g46
Award Number: 84379

Methods

Data were collected by means of face-to-face interviews with residents in each of the four towns. The residential areas of each town were first stratified into three socio-economic classes (low, low-medium and high) and the central business district. This was done on the basis of (i) scrutiny of Google Earth images, (ii) general ground familiaristion prior to data collection and (iii) experience of the researchers having worked in towns in the region. Within each of the four areas per town a target of 25 interviews was set (i.e. 100 interviews per town). However, the sample was slightly less due to logistical constraints and discarding four samples after data cleaning. The final sample was 382 participants, ranging from 92 in Kokstad to 100 in Uitenhage, and with 95 in both Mossel Bay and Butterworth. 

Within each of the four strata per town a convenience sample was employed whereby local adults (> 18 years old) were approached opportunistically in the streets but with an eye on trying to include both genders. Most interviews took approximately 30 minutes and were conducted in the language of choice of the respondent amongst isiXhosa, isiZulu, Afrikaans or English. The interview questions were read out to each respondent and their responses recorded on a paper version of the interview-schedule. The structured interview-schedule had two sections containing 20 questions in total (see Supplementary materials). The first section contained both open and closed questions on the national symbols of South Africa and the second section captured the respondent's profile using variables commonly collected in environmental and cultural heritage studies (such as age, gender, home language, highest education, income class (5 classes) and whether childhood was spent in a rural or urban area). Most of the questions in the first section focused on the national biodiversity symbols. However, a few questions were included on other national symbols (flag, anthem, coat of arms) to contextualise the responses about biodiversity symbols. The first section also asked respondents whether they felt the current symbols were appropriate. If they felt that one or more of them were not, they were invited to suggest alternative species. They were also asked whether they thought having national biodiversity symbols was useful or important. For the second question each respondent was presented with five sheets, one per taxonomic group, each with four photos one of which was the national species and three were not. They were then asked to identify the relevant national species on each sheet. Their response was recorded as either correct, incorrect, or they could not (or declined) to select one of the four images. Written informed consent was secured from each participant. Ethical approval for the study and questionnaire was provided by the Rhodes University Human Ethics Committee (2021-5095-6135).

Data analysis

The raw data were entered into MS Excel for cleaning and basic summarisation. Four samples were discarded due to incorrect data capture on the hard copy form. The data were then imported into R Studio for statistical analysis. Data normality was assessed using Shapiro-Wilk's test which indicated it was not normal. Consequently, differences between the number of national biodiversity symbols known by income group were tested using a Kruskal-Wallis test. To answer the last research question we used a binominal logistic regression to determine if any respondent attribute variables (such as age, gender, education, and including location) influenced respondents' ability to name the national biodiversity symbols. Independence of the errors and linearity were confirmed by plots of the residuals.