Survey Support Desk/Glossary
This page is currently a draft. More information pertaining to this may be available on the talk page. Translation admins: Normally, drafts should not be marked for translation. |
Participation edit
Total participants edit
Total participants includes everyone who SAW the question, whether they responded or not.
Answered question edit
Answered questions includes everyone who selected any option.
Participants in graph edit
Everyone who is represented in the graph. For scale questions or single-choice questions (e.g. Strongly agree/strongly disagree scales), the count only includes the number of people who responded in the scale. If someone chose "No opinion" or "I don't know" they are excluded. The reason for this is
Missing data edit
There are generally two types of missing data with surveys. People who skipped the question and people who's response is out of scope of the question. An example of this is "no opinion" on strongly agree/disagree scales. This response is important to the question, but it tells you who doesn't really care much about the issue or perhaps doesn't feel knowledgeable to answer the question.
Skipped question edit
The count in here are those who saw the question but chose not to respond. The only way we know this number is because Qualtrics records it for us.
Dropped out edit
This number is not yet available and likely won't be this year. Since surveys are often multiple questions, it is good to know how many people did not finish to get a sense of the completion rate.
Statistics edit
Normal distribution edit
A distribution is normal if the data look like a standard bell curve. If the data in a question is normal, this tells us that we are able to conduct additional analyses of the data.
Mean edit
The mean is only displayed if the data is normal. Otherwise the median is shown.
Median edit
When the data is not normal, or if there is a low response rate, the median is a more meaningful representation of the central tendency of the results. Comparing the Median and Mean can you information about the skew of the results.
Mode edit
The mode tells us which option was chose the most.
Margin of error edit
The margin of error only applies to questions that were found to have a normal distribution. Margin of error tells you how much error the question may have in reporting a number. For example if a mean is 3.5 and the margin of error is 10%, this means that the true value may vary from 3.0 to 4.0.
Standard deviation edit
The standard deviation, like the mean, median and mode, tell us more information about the distribution of the data and its variance (e.g. how spread out the data is).