Implementation of the quantitative survey

In order to test hypothesis H1 and hypothesis H2, a quantitative questionnaire was used. Based on the research results to be achieved, a model for hypothesis testing was developed (see p.

Figure 20). This model is based on the findings of the literature review and the results of the qualitative research.

Since the method choice was an online questionnaire, the standardized questionnaire was created using the “QuestionPro Survey“. This offers the advantage that within a short time, the achievement of large sample sizes are possible since one can distribute the access to the questionnaire by an individually created link by digital technologies (e.g., eMail, social media, messenger systems) (Bortz & Döring, 2006, p. 261).

Before the questionnaire was sent out, it was subjected to a pretest. A pretest is helpful to determine whether the questions are comprehensible and estimate the actual time required to complete the questionnaire (Ritschl, Weigl, & Stamm, 2016, p. 174). Care must be taken to ensure that the individuals completing the pretest match the characteristics of the defined population (Berger-Grabner, 2016, p. 114). In the present case, the pretest was sent to 6 people with a request for feedback regarding the questions and recorded the length of time. Since the questionnaire was distributed through “snowballing“ (Bortz & Döring, 2006, p. 128), the pretest subjects were also randomly selected. Attention was paid to equitable gender distribution, age, different levels of education, and jobs. The relevant feedback from the pretest was incorporated into the questionnaire.

Figure 20: Influence of tourism impact on the perceived quality of life

Source: Own research, 2021

The questionnaire was distributed on April 01, 2020, via email, social media, and messenger services. Professional and private contacts of the author were selected, and forwarding was requested. Employees and students at two Universities of Applied Sciences, FH JOANNEUM and FH Burgenland, were addressed as large distribution groups. Graduates of the Institute of Health and Tourism Management were also contacted. In addition, people with a high multiplier effect were explicitly asked to help in distribution (e.g., works councils, managers, association leaders). The questionnaire went offline on April 23, 2020, at 20:00.

4.2.2.1. Measurements

Following Kim et al. (2013) and Mathew and Sreejesh (2017), a survey instrument was developed whose items have been widely tested in numerous tourism impact and quality of life

2008; Sirgy & Cornwell, 2001; UNEP & WTO, 2005; Uysal, Perdue, & Sirgy, 2012). For this reason, validity tests (e.g., factor analyses) were not conducted to determine validity measures (Bortz & Döring, 2006). Cronbach’s alpha coefficient (α) for internal consistency was used to test for reliability, with  0.6 ≤ α < 0.7 indicating acceptable, 0.7 ≤ α < 0.8 good, and α ≥ 0.9 indicating excellent internal consistency (Streiner, 2003). The questionnaire consisted of three parts.

Part 1 surveys the relationship between tourism development in a community and the perceived economic, socio-cultural, and ecologic influence on the community's sustainable development (exogenous factors). The economic influence was measured by six items and was rated on a 4-point Likert scale. Answers ranged from “I fully agree” to “I do not agree”. The variable Perceived economic impact of tourism was calculated by averaging those six items. The Cronbach’s α coefficient is 0.627. The socio-cultural influence was measured by five items and was rated on a 4-point Likert scale. Answers ranged from “I fully agree” to “I do not agree”. In addition, one item was measured by giving alternative answer options. This item was examined to break up answering routines but was finally not used for statistical testing. The variable Perceived socio-cultural impact of tourism was calculated by averaging the five Likert scale-based items. The Cronbach’s α coefficient is 0.308. The ecological influence was measured by six items and was rated on a 4-point Likert scale. Answers ranged from “I fully agree” to “I do not agree”. The variable Perceived ecological impact of tourism was calculated by averaging those six items. The Cronbach’s α coefficient is 0.698. The construct Satisfaction with tourism was calculated as follows. A composite score for each of the three domains (1) Perceived economic impact of tourism, (2) Perceived socio-cultural impact of tourism, and (3) Perceived ecological impact of tourism (ranged between 0 and 4) and a total score for the variable Satisfaction with tourism, which also ranged between 0 and 4, were calculated. The scores (1, 2, 3) were weighted equally.

Part 2 of the questionnaire surveyed 20 items influencing the personal subjectively perceived quality of life (endogenous factors). The following areas were examined: (1) material satisfaction (4 items), (2) social satisfaction (5 items), (3) spiritual-cultural satisfaction (2 items), (4) satisfaction with nature (2 items), (5) sense of security (1 item), (6) satisfaction with leisure (2 items), and (7) general well-being (4 items). A 4-point Likert scale queried nineteen items (“I am not satisfied with it”; “I am rather not satisfied with it”; “I am partly satisfied with it”; “I am satisfied with it”). In addition, one question targeted the individual quality of life status compared to the personal neighborhood. Here a 5-point Likert scale was used. This item was examined to break up answering routines but was finally not used for statistical testing.

The construct Perceived quality of life was calculated by averaging 19 scores. All scores were weighted equally. The Cronbach’s α coefficient is 0.824.

Part 3 of the questionnaire surveys demographic data (Age, Gender, District of residence, Belonging to the region in years) and socio-economic data (in general and earnings from tourism). Attention was paid to the greatest possible flexibility regarding the evaluation options.

Therefore, “age“ and “belonging to region in years“ were recorded as freely entered numbers.

During the evaluation, categories can thus be formed without impairing the quality of the data in advance. Data on general socioeconomic status were collected with questions on “highest completed education“, “current occupation”, and “activity in current occupation“. The questions on education, occupation, and activity are based on the stratified sociological approach and allow assessing subjective well-being based on empirically proven determinants of health (Richter & Hurrelmann, 2009, p. 19). The ISCO (International Standard Classification of Occupations) occupational coding index has proved its worth as an instrument for surveying socio-economic status and making it comparable. The data collected from this index can be converted into the ISEI (International Socio-Economic Index of occupational status). ISEI enables high international comparability of socio-economic status - the higher the ISEI, the higher the status of a person (Züll, 2015). The last question of the questionnaire concerned the share of personal gross income generated by tourism. One of the following options could be selected by a single-choice response: 0%; 25%; 50%; 75%; 100%. The absolute figures of Social-economic status, Belonging to the region in years, and Earnings from tourism served as predictor variables.

The questionnaire can be found in the appendix (see Appendix 1, p. I).

4.2.2.2. Statistical analysis

All continuous variables were expressed as mean ± standard deviation and categorical variables as frequency (%; unless otherwise stated).

In order to test hypothesis 1 (H1), a multiple linear regression was applied. The score Satisfaction with tourism was included as a dependent variable in the model. The demographic variables Belonging to the region in years, Socioeconomic Status and Earnings from tourism, were used as predictor (independent) variables. If several independent variables are compared with a dependent characteristic, several predictor variables and one criterion variable are compared (Bortz & Döring, 2006, p. 512).

In order to test hypothesis 2 (H2) again, a multiple linear regression was applied. The score

variables Perceived economic impact of tourism, Perceived socio-cultural impact of tourism, and Perceived ecological impact of tourism and the Socioeconomic Status were used as predictor (independent) variables.