Crisis planning in the nonprofit sector: Should we plan for something bad if it may not occur?

Southern Business Review, Spring 2002 by Spillan, John E, Crandall, William

Data Collection

The survey instrument was mailed to 980 nonprofit organizations (NPOs) across the Northeastern U. S., Pennsylvania, and New York. Each survey contained a stamped, return envelope and was addressed to executive officers of each organization. One hundred and ninety useable surveys were returned, for a response rate of 19.4 percent. Results

Participants

Table 1 lists the sizes of the respondent organizations in terms of number of employees. One hundred and twenty-seven organizations (67.2%) had between 1 and 100 employees. Thirty organizations (15.9%) had between 101 and 200 employees. Twelve organizations (6.30) had between 201 and 300 employees while five organizations (2.6%) reported more than 300 employees but fewer than 400. Eleven agencies (5.8%) reported more than 400 employees. Five organizations did not respond to this question on the survey.

Table 2 lists the types of nonprofit organizations that were included in the study. Fifty-nine (31.2%) NPOs indicated their businesses were healthcare related. Forty-five (23.8%) NPOs listed social services as their agency type. Forty-three (22.890) NPOs reported other types of nonprofit areas such as cultural arts, criminal justice, and nonmental health counseling. Twenty (10.6%) NPOs reported that their focus was in education. Sixteen (8.5%) NPOs stated that they operated mental health agencies while five (2.6%) NPOs reported their service was in the child welfare area.

Hypothesis I

The first analysis examined the mean differences in the respondents' degrees of concern in those organizations with crisis management teams versus those without teams. Table 3 lists the different potential crises in descending order by t-value. Fifty-three respondents indicated that their organizations had a crisis management team while 134 said they had no such teams. Only two out of the 26 crises displayed statistically different mean degrees of concern (for a p-value of less than .05). Internet site disruption displayed a mean of 2.38 for organizations with crisis management teams. However, for those organizations with no such team, the mean was only 1.83 (t = 2.551, p = .012). The other crisis, snowstorm, displayed a mean of 3.58 for organizations with crisis management teams. For organizations without a team, the mean was 3.11 (t = 2.209, p = 0.028). For the remaining 24 crises, the means were not statistically different, thus indicating that the degree of concern was not different regardless of whether the respondents had a crisis management team. These results indicate little support for Hypothesis 1.

Hypothesis 2

The second analysis examined the differences in mean degree of concern for each potential crisis depending on whether the event had occurred at the respondent's organization. Respondents indicated that they had a higher degree of concern for 22 of the 26 crisis events if the crisis had actually occurred within the past three years (with a p-value of less than .05).

Table 4 lists each of the crises by descending t-value. Theft of company property displayed the largest difference in mean degree of concern. Seventy-nine respondents reported that they had had this crisis occur in their organizations while 108 said it had not occurred. The mean degree of concern was 3.82 for those who had had this crisis occur versus 2.08 for respondents who said that it had not occurred in their organizations (t = 10.066,