Survey Glossary

Here are some institutional terms used in the market research field:
  • Acquiescence Bias

    Acquiescence bias is the bias in research data that results from the tendency for respondents to agree with research questions or statements.

  • Cluster Analysis

    Cluster analysis is a data reduction method that focuses on respondents as opposed to variables. It reduces a large pool of respondents into groups (or clusters) based on common themes. The members of each cluster group are more similar within a group than they are between groups.

  • Conjoint Analysis

    Conjoint Analysis is a type of analysis based on a Conjoint module in a quantitative survey. The Conjoint module asks respondents assess their interest in a wide range of variations on a particular product or service offering. The analysis of this data allows researchers to identify which product or service offering will have the widest share of preference among consumers. This often includes data on how the ideal product should be priced.

  • Crosstabs

    A crosstab is a frequency table that shows data for more than one variable. This view of the data shows how groups of respondents differ in how they answer survey questions, which can help identify patterns in the data and isolate the key takeaways from a study.

  • Dichotomous Questions

    Questions that can only have 2 answers ‐ the most common type of a dichotomous question is a yes/no question.

  • Double-Barreled Questions

    Questions that ask for information on more than one question (or have answers that provide information on more than one question).

  • Factor Analysis

    In situations where many variables are at play, factor analysis identifies the key underlying factors or constructs within a set of data. A factor analysis may start with twenty known (or “manifest”) variables from a questionnaire and reduce them to, for instance, five constructs, or “latent” factors.

  • Implicit Alternatives

    An implicit alternative is an alternative alluded to, but not explicitly articulated or defined, in a question text or statement.

  • Incidence Rate

    Incidence rate refers to the frequency of something occurring within the population. The smaller the incidence rate, the more difficult it is to reach the audience.

  • Kano Analysis

    Kano analysis measures customer perceptions of product or service features, and classifies them based on satisfaction with the product according to the presence or absence of a feature.

  • Likert Scale

    A Likert scale is a widely-used rating system for survey research, consisting of a symmetrical scale of positive and negative responses. They are valuable because they streamline researchers’ ability to quantify results.

  • Margin of Error

    Margin of error is a statistic used whenever a population is incompletely sampled. It describes estimated likelihood that the result of a survey is close to the result had the entire population been surveyed.

  • MaxDiff

    MaxDiff is a form of analysis where survey respondents are shown a subset of possible items, often three to five at a time, and asked to indicate the two elements of the subset shown that are the most and least important to them. This forces respondents to make choices between options, in contrast to standard rating scales.

  • Monadic Testing

    Monadic Testing is an experimental design that allows researchers to create a single testing instrument that systematically assesses multiple different concepts and controls for the potential for bias.

  • Multiple Regression

    Multiple regression is an extension of simple regression that is used to predict the value of an unknown variable based on the value of two or more known variables. This model uses two or more independent (or “exploratory”) variables to determine a single dependent (or “criterion”) variable.

  • Mutually Exclusive, Collectively Exhaustive

    Mutually Exclusive, Collectively Exhaustive (MECE) refers to the concept of organizing information such that it is separated into buckets that are both mutually exclusive, meaning nothing falls into more than one category, and collectively exhaustive, meaning that everything falls into one of the available categories.

  • Nonresponse Bias

    Nonresponse bias is the type of survey error that stems from respondents not taking the survey. Factors including lifestyle, personal values, demographics, etc. can be noticeably different among survey respondents and all other potential respondents who chose not take the survey, and these differences bias the survey data.

  • Order Bias

    Order bias is the bias in research data that results from the tendency for respondents to click items shown at the top of a list, as these are the first items they see.

  • Randomization

    Randomization is used to remove or reduce serial position biases that can result in over or underrepresentation of certain answers based on their position within a list.

  • Rating Scales

    Rating scales are designed to quantify the presence and magnitude of attitudes and emotions. In their most traditional form, Likert scales are a 5-point scale, where 1 is the lowest point on the scale, and 5 is the highest point on the scale.

  • Regression Analysis

    Regression analysis is a technique used to understand the relationship between two variables. If there is a strong relationship between the two variables, it can be plotted on a line and the resulting formula can be used to predict values of the dependent variable. Regression analysis can reveal how closely two factors are related, as well as how certain one can be about the resulting predictions.

  • Response Error

    Response Error occurs when respondents do not provide accurate answers to survey questions. It is impossible to entirely eliminate this type of error, but careful consideration during the survey design process can help mitigate its effects.

  • Segmentation

    Segmentations are a popular market research solution for companies seeking to better understand their consumer audience by isolating how needs and behaviors differ among the consumers who buy their product.

  • Semantic Differential Scale

    Semantic Differential Scales are used to quantitatively measure attitudes and perceptions regarding people, brands, activities, etc. A typical semantic differential scale is a 7-point scale, with a neutral point in the middle. On either end of the scale are opposing adjectives or statements, so respondents select the point on the scale indicating which side of the scale best reflects their opinion, and the degree to which they feel that way.

  • Social Desirability Bias

    Social Desirability Bias is a type of response bias that occurs when respondents feel pressured, either internally or externally, to provide a socially desired response. Thus, the data from these respondents is biased and does not accurately reflect the target population.

  • Survey Fatigue

    Survey fatigue is the fatigue on respondents’ cognitive abilities that results from taking a survey. Surveys with multiple grids and/or free responses will have poorer quality answers as respondents progress through the survey.

  • Survey Logic

    Survey Logic controls how questions within a survey relate to one another, allowing a respondent’s answer to a question to affect other questions they will or will not be shown.

  • Typing Tools

    Typing Tools are spreadsheet files that allocate respondents to a particular segment based on their answers to the Segmentation questions.

  • Unstructured Questions (Also Known As Free Response)

    Unstructured questions ask respondents to provide answers in their own words into a text box. The data from these questions is classified as text, not numbers, so slight nuances in the spelling and formatting of responses results in the responses being classified as different answers.

  • Weighting

    Weighting is a data processing technique that allows researchers to change the value of a particular respondent’s answers relative to the rest of the responses in a dataset. This allows researchers to correct for sampling imbalances by adjusting the data to be more representative of their desired target population.