The standardisation and validation of a scale to measure the arts' contribution of arts festivals
Pretorius, Susanna Cornelia
MetadataShow full item record
The primary goal of the study was to standardise and validate a scale to measure the arts’ contribution of arts festivals. To achieve this goal, four objectives were formulated. First, to study arts’ contribution of arts festivals by defining, analysing and exploring related concepts through the provision of a detailed background and discussion on the topic by means of a literature review. Second, to study and select applicable explanatory theory that can be used to develop a standardised scale that measures the arts’ contribution of arts festivals by means of a second literature review. Third to determine the validity and reliability of the scale measuring the arts festival’s contribution to the performing arts and to the visual arts through an exploratory factor analysis and a confirmatory factor analysis. Lastly, to draw conclusions based on the research and make recommendations with regard to the standardisation and validation of the scale to measure the arts’ contribution of arts festivals. The first objective was achieved by conducting a literature study. The literature study on arts festivals and their contribution to the arts was explored by defining the term festival tourism; discussing aspects related to a festival (by defining a festival, explaining the festival’s relationship with culture and events, identifying the characteristics, types, benefits and potential problems associated with a festival, and identifying visitor motives for attending a festival); discussing aspects related to the term arts festival (by providing an overview of arts tourism, defining an arts festival, listing reasons for the establishment of an arts festival and discussing the arts present at arts festivals – through defining and classifying the arts); and identifying the contributions of an arts festival to the arts (by listing the types of contribution – educational, emotional, economic, quality, marketing, and growth and development – , discussing the purpose of contributing to the arts, listing perceptual differences of festival visitors regarding the contributions of an arts festival to the arts and discussing limitations of an arts festival to contribute to the arts). Achieving the second objective, the second literature study discussed the term survey (by defining a survey, identifying the types, characteristics, advantages, disadvantages and merits of a survey and discussing the survey process); discussed the term standardisation (by defining standardisation and stating the purpose for the standardisation of a measurement scale); discussed the standardisation process (by outlining the different stages in the standardisation process – the planning phase, where the aim of the measure is specified, the content is defined and the test plan is developed; the item writing phase, includes the writing and review of items; assembling and pre-testing the experimental version of the measure, which includes the arrangements of the items, finalisation of the length, protocols for answering, development of administration instructions and pre-test of the experimental version of the measure; the item and data analysis phase consists of the determination of discriminating power, preliminary investigation into item bias and the establishment of validity and reliability; and revising the final version of the measure, which encompasses the revision of the items and test, the selection of items for the final version of the test, the refinement of administration instructions and score procedures and the administration of the final version of the test); and identified and discussed ethical considerations, foreseen problems, limitations and recommendations associated with conducting a survey and the standardisation of a measurement scale. The third objective was to determine the validity and reliability of the measurement scale. This objective was achieved by discussing the implementation phase of the measurement scale where attention was given to the survey design and sampling. A stratified random sampling method was used at three selected arts festivals in South Africa, the KKNK, Innibos and Vryfees, where a descriptive survey design was administered in the form of a measurement scale, such as a questionnaire. Research assistants were trained and the scale was administered in a consistent fashion which supports internal reliability. The same survey procedures were undertaken at all three arts festivals, supporting face validity and internal validity. Representative samples were collected at the three arts festivals, also supporting external validity. Attention was also given to the measurement scale design (where respondents could give their perceptions concerning the contributions made by the arts festival to the arts through the completion of the questionnaire – which was based on literature and contained the relevant information to collect problem specific information, supporting content validity and construct validity. The scale had also undergone the delphi-technique for expert advice, supporting face validity. Providing a summary of the data analysis procedure contributed to the achievement of this objective. The data collected from the measurement scale have been captured in Microsoft™ Excel™ and analysed using the statistical software program, SPSS. The data of the arts festivals contributing to the performing arts were analysed separately from the data of the festivals where they contribute to the visual arts. This was done to get a detailed data analysis for the standardisation of the measurement scale. The split of the data also contributed to the measurement scale being divided in two separate standardised scales in determining the arts festival’s contribution to a specific form of the arts; both the data of the performing arts and the data of the visual arts had undergone the same statistical procedure for data analysis in determining the validity and reliability thereof. The results indicated that the measurement scale is a valid and reliable measure in determining the arts festival’s contribution to both the arts forms. To determine the validity of the measurement scale pertaining to the performing arts and to the visual arts, an exploratory principal axis factor analysis with Oblimin rotation was conducted on the combined data of Innibos and Vryfees (n = 982). Bartlett's Test of Sphericity was p < 0.001 and the Kaiser-Meyer- Olkin test of sample adequacy rendered a value between 0 and 1, indicating that the sample sizes were adequate to conduct an exploratory factor analysis on the data of the performing arts (KMO = 0.958) and on the data of the visual arts (KMO = 0.972); all items of the performing arts (22 items) and of the visual arts (22 items) loaded on a factor with loadings greater than 0.2. The factor analysis on the performing arts data extracted five factors. Factor 1 (Quality and Education Contribution) was defined by 5 items with a Cronbach’s α-value of 0.867 and an interitem correlation mean of 0.568. Factor 2 (Growth and Development Contribution) was defined by 6 items (Cronbach’s α = 0.896; inter-item correlation = 0.594). Factor 3 (Emotional Contribution) was defined by 3 items (Cronbach’s α = 0.706; inter-item correlation = 0.440). Factor 4 (Economic Contribution) was defined by 4 items (Cronbach’s α = 0.824; inter-item correlation = 0.540). Factor 5 (Marketing Contribution) was defined by 4 items (Cronbach’s α = 0.866; inter-item correlation = 0.617). The factor analysis on the visual arts data extracted four factors. Factor 1 (Education and Growth and Development Contribution) was defined by 8 items (Cronbach’s α = 0.947; inter-item correlation = 0.690). Factor 2 (Economic and Quality Contribution) was defined by 7 items (Cronbach’s α = 0.920; inter-item correlation = 0.622). Factor 3 (Emotional Contribution) was defined by 3 items (Cronbach’s α = 0.828; inter-item correlation = 0.616). Factor 4 (Marketing Contribution) was defined by 4 items (Cronbach’s α = 0.905; inter-item correlation of 0.704). There were correlations between factors of the performing arts and between the factors of the visual arts, where all correlations were 0.3 and larger, supporting construct validity. Further validity of the measurement scale was determined by a confirmatory factor analysis (CFA) on the performing arts data and on the visual arts data of KKNK (n = 602), where the path diagram confirmed the factor structures of both the performing arts and visual arts exploratory factor analyses, supporting criterion validity. CFA goodness-of-fit indexes were also used to determine whether the models fit with the data. The performing arts model and the visual arts model were found to have an adequate to good fit with the data of KKNK. The chi-square test of Independence (X2) for the performing arts rendered a value of p < 0.001 and for the visual arts was p < 0.001. The chi-square divided by the degrees of freedom (X2 / df) for the performing arts was 4.284 and for the visual arts was 4.9, the Comparative Fit Index (CFI) for the performing arts was 0.914 and for the visual arts was 0.931, and the Root Mean Square Error of Approximation (RMSEA) for the performing arts was 0.074 and for the visual arts was 0.079. To determine the reliability of the measurement scale, the Cronbach’s Alpha values and inter-item correlations between the factors were determined. All the factors of the performing arts rendered a high Cronbach’s Alpha value (greater than 0.7) and for the visual arts a value of 0.8. All the factors of the performing arts rendered a high interitem correlation value (greater than 0.4) and for the visual arts a value greater than 0.6. The final objective, to draw conclusions and make recommendations based on the results of the study, indicated that this study made a significant contribution to the literature and methodology of standardising a measurement scale and to the planning of arts festivals as it would lead to the development of arts festivals contributing to the arts more effectively and more efficiently. Future research on this topic should be conducted at other arts festivals, including Englishlanguage arts festivals, to enable comparative studies to be made and supporting the test-retest reliability theory on the standardised measurement scale. It is also recommended that the study should measure contribution to the arts by other arts-related organisations, for example, at museums, theatres, and galleries, by administering the standardised scale to measure the contribution they make to their specific arts form. It is important to standardise a measurement scale for arts contribution to better understand the contributing factors of the arts festival to the arts which will assist festival managers in implementing strategies that ensure the livelihood and ongoing contribution of arts festivals to the arts.