# Data aggregation for organizational culture. I calculated intra-class correlations

# ICC (1) and ICC (2) to assess the degree to which organizational culture can be

considered an organizational level phenomenon as measured in the current study. ICC (1)

can be interpreted as a measure of the proportion of variance explained by organization

membership or as the degree to which ratings from respondents from the same unit of

analysis (e.g. the organization) are substitutable (Bliese, 2000). The formula for

computing ICC (1) is based on a one-way ANOVA with organization membership as the

independent variable and the scales of interest as the dependent variable. The

organizational culture ratings on each of the four dimensions are the outcome variables of

interest. The computations of the ICC (1) an (2) were performed using the appropriate

ANOVA method.

The formula used to compute the ICC (1) is as follows:

# ICC (1)=(MSB-MSW)/[MSB+(k-1)*MSW]

# Where k is the average number of respondents from the same organization (k=2.92 in this

sample). All of the F-tests associated with organization membership were significant (p<

# .001). The ICC (1) values are as follows: .26 for clan culture, .27 for the entrepreneurial

culture, .21 for the market culture, and .43 for the hierarchy culture. These values suggest

that membership in a specific organization explains between 27% and 44% of the

variance in organizational culture ratings. These values are higher than the median ICC

(1) of .12, which was reported by James (1982), indicating that there is a good amount of

between-organization variability relative to within-organization in respondents’

perceptions of their organization’s culture.

69