How to analyze it?
To analyze the data of a survey research, it is important to:
- Determine how responses will be coded in the database. Some survey questions may be worded so that a given response represents an unfavourable rating for one question, but a favourable rating for another. In order to compare or aggregate these survey responses, the inconsistent survey question score should be reversed. To do this, switch the highest and lowest numerical values of a response, then substitute the next highest and lowest values, and so on.
- Calculate the response rate by dividing the number of people who submitted a completed survey (80% or more of the questions answered) by the number of people contacted to complete the survey. Consider other formulas for calculating responses rates, such as counting partially completed surveys as responses.
- If the response rate is below 70 percent, determine if the sample is representative of the target population by comparing sample and target population means for characteristics that would likely affect responses, such as race, age, grade point average. Large deviations of the sample from the population on demographic variables will be a signal that the sample is not representative and, thus, generalization is dangerous.
- Analyze the psychometric properties of the instrument. If different items are used to measure the same general construct (e.g., risk perception) and one wants to aggregate them into one general measure (e.g., risk perception scale), the internal consistency of this measure must be evaluated. This will help determine if the items produce similar scores, suggesting that in fact they are measuring a similar construct. Internal consistency is usually measured with Cronbach's alpha (value of 0.7 or greater indicates acceptable reliability).
- Aggregate items that measure the same construct (if they show acceptable internal consistency). This is usually done by averaging the scores of items, after their psychometric properties were assessed.
- Choose the appropriate method to analyze survey data. This decision should take into account:
The objectives of data analysis (i.e., description, relationship, prediction, comparison)
The number of independent and dependent variables
The type of scale of the independent and dependent variables (i.e., nominal, ordinal, or numerical)
- Determine if the level of significance reached is “acceptable” to reject the hypothesis that the obtained difference or correlation is due to chance. By convention, if the level of significance reached is below 5 percent or less, the hypothesis that the differences or correlations obtained are due to chance is rejected.
- Analyze open-ended questions by using qualitative data analysis, such as thematic or content analysis.