Post EE Survey ----- What to do Now?

[Clay Jordan, PHR]

I am in the process of having the EE Survey I have sent out to our base of 200+ employees returned to me through Survey Monkey. I am wanting to find out from some of you guys what you recommend be the best direction of travel after analyzing and digesting the results of this survey.

Do I develop new programs? procedures? I am a rookie at this, so jsut wanting some insight from you engagement professionals.

[Gary Cattermole]

Not sure what type of organisation but if you need external comparisons, happy to help.

Gary.

[Jean Douglas, CMC]

Hi Lydia,
While it may seem the logical way to proceed, lowest scores do not automatically = most important to work on. That is why we now see the use of predicitve modeling in analysing data. If you send me a copy of your survey, I can take a look at it and perhaps be able to give you some ideas for analysis that will lead to looking at the items that will best serve your population.

my email is [e-mail]

I hope that you are setting a minimum number of responses (10 or more respondents is the standard cut off for small organizations) to protect individual confidentiality. You will also need to read the verbatim responses, if there was an open question as they may lead to the identification of an individual - even thought that person may not be the respondent themself. This information should not be widely shared - especially if it seems likely that anyone might be able to identify the person by their comments or style of phrase.

Another caution on analyses - you will need to conduct statistical tests (e.g., t-tests or anova's) to determine if the differences between scores are "statistically significant" and then add a practical eye to the results to determine if they are of a meaningful difference (for example, two groups with scores of 73% and 75% may be statistically significant but from a practical viewpoint there is really no big difference. With small numbers that you will have there may need to be a 10 percentage point difference or more before you can say there are any differences - eg., group A has a 70% score and Group B has an 82% score but you cannot say Group B is better if the statistical tests do not confirm this ).

Another caution is "looking for differences" - you should have identified the differences you wish to explore before the analysis phase or you run the risk of hunting down a result to confirm a theory - e.g., looking for something bad until you find it can lead to a spurious result. It is okay to check out other analyses if the ones you have completed lead to more questions. I hope this makes sense. I'm not sure that I have explained this bit very well.

contact me if you have questions.
Jean

[Paul Marciano]

Programs will not lead to long-term change in employee engagement. Check out just released: "Carrots and Sticks Don't Work: Build a Culture of Engagement with the Principles of RESPECT" by McGraw-Hill.

[Ben Simonton]

Clay,

If you want help, I am willing to provide same. Please take a look at
http://www.leadingtoday.org/Onmag/jan03/bs-jan03.html

If you are interested, call me at 813-634-4003

Best regards, Ben