| 4.11.2 Reliability
The reliability of the measurement instrument is concerned with the consistency of the responses to the questions (Mitchel, 1996). Reliability is the accuracy or precision of the measurement instruments such that it would be able to yield similar outcome or results for similar inputs. Statistically, reliability is defined as the percentage of the inconsistency in the responses to the survey that is the result of differences in the respondents. This implies that responses to a reliable survey will vary because respondents have different opinions, not because the questionnaire items are confusing or ambiguous.
It could be estimated using stability or equivalence approaches. In this study we pre-tested the questionnaire to strengthen its reliability. Although our data measuring instruments are in the form of dichotomous choices (Yes/No) the Cronbach’s alpha coefficient was used because it will produce the same results as the Kuder-Richardson Formula (KR-20) with dichotomous data (Cortina, 1993; Yaffee, 1999). Cronbach's alpha, developed in 1951 (Cronbach 1951) is a model of internal consistency based on the average inter-item correlation. The Cronbach's alpha is calculated based on the number of items on the survey (k) and the ratio of the average inter-item covariance to the average item variance.
This ratio assumes that the item variances are all equal; it specifies to the average inter-item correlation, and the result is known as the Standardized item alpha (or Spearman-Brown stepped-up reliability coefficient):
A summary of the results of the Cronbach’s alpha retrieved from the SPSS output for all the five marketing dimensions is shown in table 4.6 below:
| 