On a regular basis I get questions about how we are going to ensure security of survey data. The majority of our work involves employee surveys, and with these surveys I would say don’t use a password to enter the survey.

1. Candid feedback is anonymous and a password entrance leaves the impression that participants may be tracked.
2. We ask questions about trust of senior leadership all the time and using passwords undermines this message.
3. Every step to participate in a survey lowers the response rate.
4. We can track the responses coming in and track if “ballot stuffing” has occurred without using passwords.
5. We use a simple cookie to discourage multiple submissions.

Offering rewards for participation encourages multiple submissions. The value transaction for employee surveys should be the impact the results will have them. The key to validity is to encourage ownership of the process and the results.

Passwords may need to be used

Public internet surveys may need different levels of security. I have a good friend that put together a survey on a local chamber of commerce website. It was an issue survey. To publicize the survey he sent out emails to his contact list. His hope was to offer support to his cause from the business community. Unfortunately for him the results were unrealistically supportive and completely useless. I reminded him to call me the next time he thought of doing a survey.

In his case, they needed the chamber to run a secure survey to just the Chamber members. In this case a password would be important to maintain audience control. In community relations issues political activism can cause your data to be at risk.

How we protect our data is situation dependent, but we must also look at how we do the research communicates to our audience. What has been your experience around securing employee survey data?

Completely unrelated.
I came across a great study on linking employee moral with stock price.