Higher education system governance: an exploratory study of presidential decision making in the kentucky community and technical college system

Quantitative Data Collection

An email generated through Qualtrics software was sent to the campus email addresses of these participants in late March 2016. The email explained the purpose of the study, provided a description of the survey, explained procedures used to ensure confidentiality, indicated how long the survey will take to complete, and indicated that completion of the survey indicated voluntary consent to participate in this phase of the study. The email also contained the link to the survey with a requested completion date in late March 2016.

Following the initial survey solicitation email and one reminder email, 5 participants had completed the survey. After completing quantitative data collection and
commencement of qualitative data collection, the researcher received consent from 1 additional participant for phase one of the study. The survey email was sent to this participant in early May 2016. As a result, preliminary analysis reflects survey data from 5 participants. The researcher calculated descriptive statistics for survey data from all 6 participants.

Quantitative Data Analysis

Survey data were intended to identify the location of decision making for specified academic, administrative, and personnel decisions as outlined in the research questions guiding this study. Also, survey data were intended to inform the development of a semi-structured interview protocol because there are two, sequential phases to this study.

Preliminary analysis. Following administration and completion of the survey, survey data were downloaded from Qualtrics for preliminary analysis. Preliminary analysis involved calculating the number of participant responses along the scale for each of the decision items in table format, with 5 being the highest number of responses and 0 being the lowest number of responses across the scale. This illustrated where the majority of participants perceived decision making occurs for each decision item.
In addition, the researcher calculated the total number of participant responses along the scale for all decision items, which provided an overview of the dispersion of participant responses across the scale. Then, the researcher sorted the total number of participant responses along the scale by the academic, administrative, and personnel categories, which provided an overview of the dispersion of participant responses by decision category examined in this study.
Descriptive statistics. The researcher calculated descriptive statistics following the analysis plan used by Ingram and Tollefson (1996), which included analyzing frequencies and calculating measures of central tendency and measures of dispersion. Because the survey used a five-category, modified Likert scale, the data were assigned a numerical value. Table 3.3 outlines the range of possible values and assigned numerical values.
Table 3.3


Range of Possible Values for Location of Decision Making

Scale	Value
The local college	-1
Primarily the college, with some input from the state community college system	-0.5
Shared equally between the college and the state system	0
Primarily the state system, with some input from the colleges	+0.5
The state system	+1

Descriptive statistics involved analyzing frequencies at the decision-area and decision-item levels, as well as total frequencies for the scale. In addition, the researcher calculated the overall mean and the means by decision areas, as well as the range. The researcher analyzed frequencies to assess the degree to which decisions within the academic, administrative, and personnel decision areas occur at the local community college; primarily at the college, with some input from the state community college
system; are shared equally between the college and the state system; primarily the state system, with some input from the college; or, the state system.
Specifically, frequencies were analyzed for each decision item response to assess the degree to which decision making was perceived to occur at the locations indicated on the scale. A total frequency was calculated and analyzed for all items according to the scale, along with subtotals for the academic, administrative, and personnel decision areas to determine if there were differences in the perceived location of academic, administrative, and personnel decision making.
In addition, the researcher calculated the mean value for participant responses for the decision items asked about on the survey as well as a mean value for the overall location of decision making. Because numerical values were assigned to the response scale of where decision making occurs within the community college system, a mean value was calculated for each survey item to determine the degree to which decision making for each decision item is perceived to occur at the locations indicated on the scale. The total range of possible mean values was -1 to +1. A total mean value for the dataset was calculated along with subtotals for the academic, administrative, and personnel decision areas to determine if there were differences in the perceived location of decision making for these areas. Finally, the researcher calculated the range to determine whether there was dispersion in participant responses across the scale.