I have recently administered an employee survey at a call center. I found a Pearson r of -.66 between the employees' ages and their scores on an employee engagement scale. What does this finding represent/mean? Also, there is a relationship between gender and employee engagement (chi-square was statistically significant at p<.05. I do I now perform an analysis that incorporates both gender and age in predicting employee engagement? Plus how do statistical significance and substantive (or meaningful) significance differ?
Apparently, age accounts for .4356 (.66^2) of the variance of the scores.
Use chi-square with cells for age and gender.
Example: Relation between height and level of executives for males was significant, with greater height indicating higher level (statistical significance). However, the differences were only in fractions of an inch, not allowing much prediction of executive level (substantive significance).
The Pearson r value of -0.66 between employees' ages and their scores on an employee engagement scale represents the strength and direction of the relationship between the two variables. In this case, a negative correlation indicates that as employees' ages increase, their scores on the employee engagement scale tend to decrease. The value of -0.66 suggests a moderately strong negative relationship between age and employee engagement.
To perform an analysis that incorporates both gender and age in predicting employee engagement, you can use multiple regression analysis. Multiple regression allows you to examine how multiple predictor variables (such as age and gender) relate to a single outcome variable (employee engagement). The analysis will estimate the unique contribution of each predictor while controlling for the other variables.
To conduct this analysis in a statistical software package, follow these steps:
1. Choose the appropriate multiple regression analysis tool in your preferred statistical software (e.g., SPSS, R, Excel, etc.).
2. Specify your outcome variable (employee engagement) as the dependent variable.
3. Include age and gender as independent variables or predictors.
4. Run the analysis and interpret the results, considering the coefficients (beta weights) for age and gender, as well as their significance levels.
Statistical significance and substantive significance are two different concepts.
Statistical significance refers to the probability that the relationship or difference observed in the data is not due to chance. It is typically assessed through hypothesis testing and is represented by a p-value. A significant result (e.g., p < 0.05) indicates that the observed relationship or difference is unlikely to occur by chance alone. However, statistical significance does not necessarily imply practical or meaningful importance.
On the other hand, substantive or meaningful significance refers to the practical importance or real-world implications of the relationship or difference observed. It takes into account the effect size and practical context. A small but statistically significant effect may not have meaningful importance, while a large effect with practical consequences may be meaningful even if it does not achieve statistical significance.
Therefore, it is essential to consider both statistical and substantive significance when interpreting the results of your study.