The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
evaluation of survey
Contributor II
Dear All,
in our company we conducted a survey among the employees (250 employees) . The participants could answer some questions to different topics (e.g. topic A: "communication", topic B: "Own actvities", topic C: "direct superior" (= direct boss), ... )
The possible answers (nominal) were "not relevant at all" (= 1 point), ..., "applies completely" (= 5 points).
Now, I want to find groups of employees within our company (e.g. -> satisfied employees, ...)
Up-to-now, I tried k-means: I calculated the average points for each employee within each topic and ran the k-means. here I found four groups (the best number of groups).
Are there any other approaches (PCA, ...) to find groups within the results of the survey?
Thanx in advance
currant
in our company we conducted a survey among the employees (250 employees) . The participants could answer some questions to different topics (e.g. topic A: "communication", topic B: "Own actvities", topic C: "direct superior" (= direct boss), ... )
The possible answers (nominal) were "not relevant at all" (= 1 point), ..., "applies completely" (= 5 points).
Now, I want to find groups of employees within our company (e.g. -> satisfied employees, ...)
Up-to-now, I tried k-means: I calculated the average points for each employee within each topic and ran the k-means. here I found four groups (the best number of groups).
Are there any other approaches (PCA, ...) to find groups within the results of the survey?
Thanx in advance
currant
0
Answers
if you could calcuate the average for each employee, it means that you already transformed the nominal user input to numerical values (1-5 for each answer). On these values you could try a clustering algorithm like k-means *without* prior averaging the values. That way you use more information you gathered by your survey than with the averaging approach.
Cheers,
Marius