2008 Employee Survey Results - Laserfiche WebLink

Home Browse Search

City of Louisville Employee Survey July 2008 Calculating the Composite Ratings A statistical procedure known as factor analysis was conducted to identify the underlying factors of employees'work experience that could be gleaned from the survey questions.This examination identified the individual items (such as "Overall,I am satisfied with my job,""I have the right skills and abilities for doing this job," etc.) that shared a common theme such as job satisfaction or employee fit.The factor analysis resulted in 12 main factors of employee work experience.Themes or factors were identified when the"factor loadings" (a measure of contribution to the factor) for each survey question were generally 0.40 or higher and when the reliability among each identified factor was generally greater than or equal to 0.70. Reliability analysis of the resulting composite ratings was used to confirm that each factor had an acceptable level of internal consistency when the items were grouped together.This is generally measured by Cronbach's alpha,a statistic that measures the extent to which question items within a scale measure the same construct.While there are no hard and fast rules about what levels of Cronbach's alpha are acceptable,one author has proposed that levels "of 0.70 or more are generally accepted as representing good reliability" (Litwin,MS.,How to measure survey reliability and validity. Thousand Oaks: Sage Publications; 1995),while another states that"[a]s a general rule,we believe that reliabilities should not be below 0.80 for widely used scales" (Carmines, EG, Zeller,RA. Reliability and validity assessment. Newbury Park: Sage Publications; 1979).All but two of the factors had an internal consistency of 0.70 or above.The table on the following page shows the items that comprise each of the composite ratings and the factor loadings,as well as the Cronbach's alpha. Examining Significant Associations with Employee Job Satisfaction Multiple linear regression analyses were performed to determine which composite ratings were associated with overall employee job satisfaction.A multiple linear regression model allows the simultaneous examination of the association of multiple factors with a single outcome measure of interest,often referred to as the dependent variable (in this instance,overall employee job satisfaction,question#1a).The factors examined for an association with the dependent variable are referred to as independent or predictor variables.This simultaneous examination allows one to look at a particular association of interest, for example the association of employer communications, simultaneously adjusted for all the other variables in the model. Regression coefficients are calculated for each predictor variable in the model.These coefficients can be interpreted as a slope, that is, for every unit change in the predictor variable,the independent variable would change by the amount of the regression coefficient.A test of statistical significance is calculated for each regression coefficient,with a corresponding p-value.A p-value refers to the probability that the regression coefficient is significantly different than zero (meaning there is no association between the predictor variable and overall employee job satisfaction.A p-value of less than 0.10 means that the probability of no association is less than 10%. For Louisville,two of the 12 composite ratings examined were found to be strongly and significantly associated with overall job satisfaction in the multiple linear analysis:Department Communication and Policies and Employer Communication.The correlation of each of the 12 composite ratings was examined individually with overall job satisfaction.All were found to be individually significantly associated with overall job satisfaction,but in addition to Department Communication and Policies and Employer Communication, several others were found to have fairly strong associations (Pearson correlation coefficients of 0.530 or above);these included Employee Fit and Values, Growth and Training Opportunities,Job Pride and Contentment and Job Satisfaction. a DRAFT Report of Results 43