Term | Definition |
Base | The number that is used to compute percentages in a table. Typically a count of respondents who saw a question response in a survey. |
Frequency | The numerical value representing how many respondents answered a question. This can also occur within a given time period if the survey is conducted over a period of time. |
Grid Question (also known as a Matrix Question) | A question typically presented as a table where the respondent evaluates rows of options using the same column choices.
Grid questions populate both rows and columns with Variables: One variable representing the rows and one representing the columns. |
Low Base Threshold | A pre-defined attribute that highlights when too few respondents are present in the data. Typically, data that has a low base size is not used to make business decisions.
This can be important in cases where a high percentage is showing, but only 10 respondents make up the base. |
Multiple Choice Question (also known as a pick Pick Many Question) | A type of question that asks the respondent to pick as many choices from a list of options that are relevant to them. (e.g, ‘Which of the following characteristics do you look for in your candies? Pick all that apply.’ ) |
Options (Grid/Matrix) | The subcategory (brand/product etc) that differentiates the questions within a Grid |
P-Value | The P-value, or probability value, is a measure of the probability that an observed difference in data occurred by chance or sampling error.
P-values range from 0-1. The lower the P-value is the more strongly it validates that the chance/confidence of statistical significance is valid.
P-values is compared to the significance level to tell us whether something is statistically significant.
*See significance level and statistical significance |
Respondent | A person who answers one or more questions in a survey. |
Response(s) | A categorical or numeric value that a Respondent may select as part of answering a question |
Sample | The total number of respondents who took a survey. |
Significance Level | A value used to measure the likelihood that our testing hypotheses are true.
For example, if your significance level is at .05 (95%), you're saying that there is a less than 5% chance that the results occurred by chance or sampling error. We can feel confident with these results.
Lower the significance level, lower the confidence in the validity of the data telling you it did not occur by chance.
*See also statistical significance |
Single Choice, or Pick One Question | Type of question which asks the respondent to pick one choice from the provided options. (e.g, ‘What is your favorite color?) |
Statistical Significance | Statistically significant data indicates a relation with other data that is not likely to be occurring by chance.
By default, KnowledgeHound sets the statistical significance level at .05. This indicates there is >95% probability the difference between populations is significant and not due to random chance or sampling error.
KnowledgeHound indicates statistical significance with a yellow highlighted indicator when comparisons have a p-value lower than the significance level. (e.g. - A comparison of females against males that has a p-value of .034 will show an indication as being statistically significant. If the p-value is above .05 you will not see any indication.) |
Total | The count of respondents who provided a response to a survey question(s). |
Variable (KnowledgeHound's definition) | A unit to represent a survey question(s) and its response(s). For example, adding a Grid Question in our Analysis page will create 2 Variables - one for the Responses, one for the Options. |
Variable (statistical definition) | A statistical term to describe a quantity with an assigned value that may change during research. |