Smart Survey Design
statistics. For a simple explanation of the difference between descriptive statistics and inferential statistics please visit the following site: http://www.statisticallysignificantconsulting.com/Statistics101.htm
Inferential statistics allow you to find out if your results are “statistically significant” or not. For example, you might develop an opinion survey and measure basic demographic characteristics of the study participants such as age and gender. Using inferential statistics, you could determine if there is a statistically significant difference in opinion between males and females, or if the opinion improves with the age of the participant.
In order to use statistical inference effectively, there are several statistical considerations that need to be taken into account when you are planning your survey. Some of these considerations are: 1.) Research questions 2.) Measurement scales 3.) Null and alternative hypotheses 4.) Data analysis plan 5.) Sample size
You may want to consult with a statistician (e.g. www.StatisticallySignificantConsulting.com) at the design phase of your study to ensure that your data and research questions will lend themselves well to statistical analysis (Creech, Steve 2007).
Survey Administration – The ways in which the surveys are administered play a role in response rates for surveys and these can be relative:
Mail: 50% adequate, 60-70% good to very good
Phone: 80% good
Email: 40% average, 50-60% good to very good
Online: 30% average
Classroom pager: 50+% good
Face to Face: 80-85% good
Closeness or Relationship to Respondents - The better you know your respondents, the more likely you will have a higher response rate.