Survey User's Guide
Appendix A. Selecting a Sample of Staff
You can administer surveys to all staff in your nursing home, or to only a subset or sample of staff. Although surveying all staff may seem simple or most desirable, it will take additional time and resources.
If you choose to sample part of your staff, your goal should be to select a group of people who closely represent the characteristics of all your nursing home staff. That way, you can be more confident that your sample's results adequately reflect the opinions of staff in the whole nursing home.
Review the following suggested steps for selecting a sample:
- Include all executive leaders, all managers, all physicians, and all staff positions where there may be only one or two staff in the position, such as dietitians and activity directors.
- Determine how many of your remaining staff you want to sample; most of these staff will be nurses, nursing assistants or aides, and other support staff. Although your resources may limit the number of staff you can survey, the more staff you survey, the more likely you are to represent your nursing home adequately.
- First, think about your budget and how many responses you want to receive (i.e., your response goal). Your budget will need to include the costs of distributing both first and second survey packets. Your budget also should include additional costs for materials such as envelopes, as well as postage if respondents will mail their completed surveys to a vendor or to a system headquarters address.
- Second, think about your expected response rate. Because not everyone will respond, you can expect to receive completed surveys from about 40 percent to 60 percent of your sample. Therefore, to reach your response goal, if you assume a response rate of 50 percent, your sample size should be at least twice the number of responses you want to receive. If the number of responses you eventually want to achieve is 100 completed surveys, be prepared to select about 200 staff.
- To select individuals to include in your sample, develop a list of all the staff in your nursing home who meet your eligibility criteria for participating in the survey. Exclude management, physicians, and other staff from Step 1 since they are already included in the list of people you will survey. We suggest using software such as a Microsoft Excel® spreadsheet that will let you sort your list. In the following example, we have a nursing home with 500 staff (excluding management, physicians, etc., from Step 1) and want to select a sample of about 200 staff to receive 100 completed surveys.
- In the first column of your spreadsheet, list the numbers 1 to 500 for every eligible staff member. In the second column, include the first and last name of each staff person. In the third column, list the person's job title. It is important that the job titles you enter for the same staff position are consistent because you will sort this column.
- Review the list to make sure it is appropriate to survey each staff member on the list. To the extent possible, ensure that your list information is complete, up to date, and accurate. Points to check for include:
- Staff on administrative or extended sick leave.
- Staff who no longer work at the nursing home.
- Other changes that may affect the accuracy of your list of names.
If you believe there are certain staff who should not receive the survey, selectively remove them from the list.
- Electronically sort the list by job title so that, for example, all licensed practical nurses are grouped together and all housekeeping/linen service staff are grouped together.
- After sorting your list, select your sample of 200 staff from the 500 on your list as follows:
- First, divide the number of staff on your list (500) by the number of staff you want to select from the list (200). The result is 500/200 = 2.5.
- When the result is a whole number such as 2, this number tells you that you should select every second person on your list after a random starting point. When the result is a fractional number, such as 2.3, 2.5, or 2.7, it will be easier to select your sample if you round the number down (to 2 in this example) and use that number to select every "nth" person on your list (in this example, every 2nd person).
- Start with a staff member chosen randomly from your list of 500 staff; then select every 2nd person on the list. For example, say you randomly chose staff member #270 as your starting point. You would select numbers 270, 272, 274, and so forth, on your list. Continue to the end of the list and then keep selecting staff from the beginning of the list until you reach person #270 again. Stop.
- Because you rounded from 2.5 to 2 in this example, you will select about 250 staff, which is 50 more than your planned sample size of 200. You can either survey the larger selected sample of 250 staff, or you can randomly remove 50 staff members from your selected sample.
This method of sample selection—sorting a list of staff you want to sample by job title and systematically selecting sample members from the list after a random start—will increase the likelihood of selecting a sample that is representative of your entire nursing home.