英语论文网

ction was done through 'mail survey and personal interviews'. The personal interviews served as a pilot study which enabled the researchers to test for the understanding of the questionnaires among a few of the intended respondents. This enabled them to refine the data collection instrument for clarification and simplicity. The process adopted by the researchers in the design of the data collection instrument is ideal for this kind of study with similar processes followed by prior and subsequent studies (Abdel-Kadar and Luther 2008; Chenhall and Langfield-Smith, 1998; Govindarajan, 1988).
Govindarajan (1988) for instance employed interview to refine the questionnaires to capture the specific objective(s) intended, reviewed prior studies to find questionnaire instruments fitting for the study, discussed the draft questionnaires with experts who help fine-tune it to give it 'content validity' and ensured a pre-testing of the instrument among the class of intended respondents to ensure clarity and understanding. Jermias and Gani (2004) have followed a similar path to giving validity to the research instrument.
However, improvements could have been made to the data collection process by involving more than one respondent per firm (in different sub units) so as to be able to compare the different responses for commonalities. Where a member of the top management has been selected to make a nomination of the participant, Govindarajan (1988) ensured a cover letter was sent to the nominated official explaining the rationale of the study and giving assurance that information given will be confidential. This has the effect of allowing the nominated personnel to be as open as possible in giving responses without the fear of what consequences my follow certain responses. The researchers failed to create that impression among respondents.

Data Analysis/ Warranted Conclusion 数据分析/保证结论

Given a high response rate to the survey and data from the interviews, the researchers found it compelling to analyze the data without taking into consideration a test for non-response. A total of 106 valid responses representing a 92% response rate were recorded. Statistically these responses were subjected to thorough analyses, first to sort the data into strategy adopted by organizations (product differentiation or low-cost). They then evaluated the concept of fit by 'calculating the weighted sum of the fitness contribution of each contextual variable using the formula derived from the fitness landscape theory…' (p. 193).
A correlation test was run to ascertain the relationship between the concept of fit and business unit effectiveness. Comparison was also made between the variable studied. A reverse coding of some of the variables analyzed was done to ensure consistency in result. Overall, a thorough analysis was conducted on the data gathered, the results of which gives insight to the measurement of contingency variables.
On the basis of the analyses of the data collected by the researchers, they asserted that the effectiveness of a business unit was dependent on the fitness between its contextual variables and the strategy adopted. Their finding upholds the concept of the contingency theory, that there are no optimal systems for organizations. Rather organizational structures leading to effectiveness are the outcome of various situational or contextual factors acting on them. Although the re