MATERIALS AND METHODS
Data collection: A cluster sampling technique was used for this study. In this method, the researchers selected the groups (agencies) and then from the elements within each selected group (customers) the required informations were collected. Therefore, from 10 Iran Khodro (The biggest car factory in Iran) car service agencies in the target population, 4 were chosen. After that, 400 customers were selected based on the Morgan table. Respondents were selected amongst the customers who had Peugeot 206, Peugeot 405 or Peugeot Persia and visited the sampled agencies during day time and at various days for a week or a month. The customers who had Peugeot 206 account for 32.5%, Peugeot 405 and Peugeot Persia make up each 33.75% of the customers. Every customer who entered the agencies handed a questionnaire.
Measures: Data were collected by using the SERVQUAL questionnaire as a measurement tool, conducted in four agencies in Iran during October to December 2007. The SERVQUAL scale is a two part instrument. First part of the questionnaire includes 22 items which measures expectations of customers in conjunction with second part consisted of 22 items which evaluates their perceptions of actual performance.
The SERVQUAL instrument was slightly modified to reflect the nature of delivery services in car agencies. The original items were translated into Persian language and the validity of content such as wording and meaning were checked carefully by two Iranian experts. A seven-point Likert scale was used to measure the expectation and perception of service quality which ranged from 1 = strongly disagree to 7 = strongly agree. A pre-test was then conducted with 20 customers and minor changes to the scales were made to ensure that the questions were not repetitive.
RESULTS
Reliability tests: To assess the internal consistency reliability of the five dimensions, Cronbach’s coefficient alpha was computed for both the expectation and perception sections. These alpha coefficients ranged from 0.71 to 0.89 for expectation part and alpha ranged from 0.65 to 0.85 for perception part. As it has been shown in Table 1; the dimensions of service quality appeared to have satisfactory reliable.
Hypothesis testing: The first hypothesis stated that there are significant differences between customers’ expectations and their perceptions of existing levels of service quality at the delivery service department of their own agencies. Differences were calculated by subtracting the mean ratings of expectations from the mean ratings of perceptions. The negative value is characterized as a service quality gap (Parasuraman et al., 1985). We performed dependent t-test to determine if the differences between customer’s expectations and perceptions were statistically significant.
Table 2 showed differences between customer’s expectations and perceptions of service quality in each agency. The negative values were obtained for all dimensions except the dimension of tangibles in all agencies.