The platform will undergo maintenance on Sep 14 at about 7:45 AM EST and will be unavailable for approximately 2 hours.
2002
DOI: 10.1002/smj.232
|View full text |Cite
|
Sign up to set email alerts
|

Measuring and testing change in strategic management research

Abstract: This paper reviews some critical issues associated with measuring and testing change and then reports on how strategy researchers have addressed those matters. We first discuss three key methodological requirements: reliability assumptions of change variables, correlations between the change variable and its initial measure, and selection of unbiased measurement alternatives. Next, we present data from a content analysis of 126 change studies which suggest that strategy researchers tend not to recognize those … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
65
0

Year Published

2007
2007
2021
2021

Publication Types

Select...
9
1

Relationship

0
10

Authors

Journals

citations
Cited by 80 publications
(71 citation statements)
references
References 22 publications
0
65
0
Order By: Relevance
“…The positive significant result for compensation to performance changes indicates that studies using change variables tend to identify stronger focal effects, which may well be a statistical artifact (cf. Bergh & Fairbank, 2002). The negative significant result for the lagged performance measurement variable suggests that pay is more sensitive to current than to past performance.…”
Section: An Institution-based View Of Executive Compensationmentioning
confidence: 83%
“…The positive significant result for compensation to performance changes indicates that studies using change variables tend to identify stronger focal effects, which may well be a statistical artifact (cf. Bergh & Fairbank, 2002). The negative significant result for the lagged performance measurement variable suggests that pay is more sensitive to current than to past performance.…”
Section: An Institution-based View Of Executive Compensationmentioning
confidence: 83%
“…We use a difference score to measure change over time, as a dependent variable, and thus these problems are not relevant to our study. Second, many change measures have low reliability as a result of high correlation with the level measures used to construct them (Bergh and Fairbank 2002). We have the opposite situation in which our measure of change in diversification has low correlation with the level of diversification (−.06 for shipbuilding, −.08 for robotics).…”
Section: Variables Diversification Changementioning
confidence: 96%
“…We often find a close relationship between the past value of a variable and its future value. The change and the past score share variance and their correlation is generally negative because the variance of future scores of a variable is often smaller than the variance of past scores (Bergh and Fairbank 2002). Given that the value of the IJV ratio at t might be influenced by its prior value at t−1, we included the prior level of IJV ratio in the models.…”
Section: Control Variablesmentioning
confidence: 99%