 |
|
Glossary of Terms
A
Analysis Of Variance (ANOVA)
Statistical test used to determine if two or more groups are significantly different in their
responses on a particular issue
BACK TO TOP
B
Baseline Research
Research designed to collect data to be used as a basis for comparison against future data
Binomial Variable
A variable with only two attributes, e.g. gender
BACK TO TOP
C
Callback
An additional telephone call made to each phone number in a sample frame that was not answered on the initial call. In a typical telephone survey methodology, interviewers will use four or five callbacks on each number until the call is completed.
Census
A count of every member of a population
Closed-Ended Question
A question to which the respondent is limited in the type of responses he or she may give, e.g. there are only three possible responses to the question, "Will you buy this product"-"Yes," "No," or "I don't know"
Cluster Analysis
A multivariate analysis technique that seeks to organize information about variables so that relatively homogenous groups, or "clusters," can be formed
Coding
A process by which verbatim comments or open-ended responses are categorized based on their similarities and are assigned numeric values
Completion
A completed questionnaire or interview
Computer Assisted Telephone Interviewing (CATI)
An interviewing technique by which the interviewer is prompted by the question on a computer monitor and enters the participant's response electronically
Confidence Interval
A range of values within which a population mean lies, for a given confidence level
Confidence Level
The probability that a population mean lies within a given range of values
Conjoint Analysis
A statistical technique in which respondents' utilities or valuations of attributes are inferred from the preferences they express for various combinations of these attributes
Consumer Behavior
The actions and attitudes of individuals or groups regarding the selection, purchase, use, and disposal of a product or service
Convenience Sample
A sample selected merely on the participants' availability or ability to participate
Correlation
An established relationship between two variables
Cross-Tabulation
A method of reporting responses to one or more variables based on the responses to other variables
BACK TO TOP
D
Database
A collection of related data organized for quick access
Demographic Data
Objective and descriptive population data that is easily identifiable, e.g. age, income, gender, and education level
Dependent Variable
Variable in an experiment that is thought to be affected by (to depend on) another variable (independent variable)
Discriminate Analysis
A statistical technique employed to model the relationship between a dichotomous or multichotomous criterion variable and a set of continuous predictor variables.
Discussion Guide
A list of issues and specific questions used by the interviewer to direct the discussion topics in a focus group or in-depth interview
BACK TO TOP
E
Empirical Data
Verifiable information that is collected through scientific observation or experiment. For example, results from surveys or focus groups are empirical data.
External Validity
The extent to which the results of a measurement of a sample can be accurately applied to the population; generalizability
BACK TO TOP
F
Factor Analysis
A multivariate analysis technique used to analyze the interrelationships among a large number of variables and to explain the relationships among variables in terms of their common underlying dimensions
Fielding
To collect data using a particular survey instrument-a survey is "in the field" or a researcher is "fielding" a survey during the actual data collection process
Focus Group
A research methodology by which 8 to 10 individuals are invited to discuss a particular issue in a small group context; focus groups are facilitated by a moderator and rely on the interaction between participants for their research value; normally, focus groups last 90 minutes to two hours each
Frequency Distribution
A report of the number and type of responses to a particular question, e.g. the following is a frequency distribution for an income variable
| Response | # of Occurrences |
| Under $30,000 | 11 |
| $30,000 to $45,000 | 31 |
| $45,001 to $60,000 | 29 |
| Over $60,000 | 14 |
| Total | 85 |
H
Honorarium
A stipend paid to research subjects as an incentive to participate
BACK TO TOP
I
Incentive
See Honorarium
Incidence Rate
The frequency of something occurring in the population. It usually refers to people, e.g. the percentage of pet owners in the population.
Independent Random Sampling
A method of selecting research participants from the population based on a random number-generating algorithm
Independent Variable
A variable in an experiment whose value is thought to affect the value of another variable (dependent variable)
In-Depth Interview (IDI)
A research methodology by which an individual discusses a particular topic at length with a trained interviewer; normally, an IDI lasts approximately 30 minutes
Internet Panel
A database of individuals who have volunteered to participate in online research studies
Interval Scale
A measurement scale in which all levels are equally spaced, such that an increase of one unit on the scale is equal to the same increase at a different point on the scale; it does not include a true zero point; e.g., the 10-point rating scale (1 to 10) used in gymnastics is an interval scale
BACK TO TOP
L
Leading Question
Questions designed to direct the respondent to provide a response desired by the questioner
Likert Scale
A five-point scale measuring a respondent's level of agreement with a given statement
Likert-Type Scale
Any measurement scale wherein a respondent is asked to rate his or her attitude regarding a given statement
Listed Sample
A sample frame drawn from a database with known values for a variety of variables used when targeting a particular segment of the population for research purposes; e.g., when conducting research on the use of the Internet among retired persons, the researcher could use a sample frame of persons age 65 years or older
BACK TO TOP
M
Mall Intercept Interview
A research methodology by which potential respondents are intercepted at the interviewer's discretion to participate in the research study; usually conducted in a shopping mall.
Margin of Error
See Sample Error
Mean
The average value
Median
The mid-point; exactly one-half of responses are less than the median and one-half are greater than the median.
Methodology
The process used and steps taken to collect data in a research effort
Metropolitan Statistical Area (MSA)
A measure used by the U.S. Census Bureau to define an urbanized area with a total population of at least 100,000
Mini-Group
A focus group conducted with only 4 to 6 participants and last for approximately 60 minutes; mini-groups are commonly conducted with teenagers or children, with hard-to-recruit populations such as community leaders, or in a context such as an online chat room or a teleconference in which communication is more limited
Mode
The most common response or value
Moderator
A trained professional who facilitates and maintains topic control during group discussion
Moderator's Guide
See Discussion Guide
Multivariate Analysis
A statistical analysis of the simultaneous relationships among three or more variables
BACK TO TOP
N
Nominal Scale
A measurement scale in which the levels can be assigned arbitrary numeric values, but the values have no intrinsic order or mathematical properties; for example, race and gender are both measured using a nominal scale
Nth Sampling
A method of selecting potential research participants by selecting every nth person in a population to participate in the sample; for example, a restaurant could ask every 10th paying customer to complete a satisfaction survey
BACK TO TOP
O
Online Focus Group
Participants simultaneously log into a private, Internet chat room hosted by the research company and provide their thoughts and attitudes (via typed comments) about the proposed topic. These sessions are guided by a professional moderator and permit clients to view the discussions in real-time from their computers.
Open-Ended Question
A question to which the respondent is not limited in the type of answer he or she can give, e.g. "Why do you prefer golf to tennis;" See Closed-Ended Question
Outliers
An extreme value that does not fit into the normal range of values for a given variable; for example, in a sample of test scores if most students score between a 75 and a 98, a student with a score of 30 would be an outlier
BACK TO TOP
P
Perceptual Map
A graphic representation of a company's or product market positioning according to perceptions of the target market
Population
The total number of people in which a marketer is interested
Pretest
A smaller version of the study conducted with 20 to 30 respondents prior to the actual data collection period; a pretest is conducted to evaluate the validity of the survey instrument and to identify any potential problems with the study
Primary Research
Research conducted with a sample of the target population
Probe
A type of follow-up question asked of a respondent to clarify or expand on the initial response; e.g. "why do you believe that," "how so," and "what else"
BACK TO TOP
Q
Qualitative Research
Research such as focus groups and in-depth interviews, the results of which cannot be statistically applied to a population
Quantitative Research
Survey research using a sample of people drawn at random from a given population. If the sample is drawn properly, the results of quantitative research can be generalized to the population
BACK TO TOP
R
Random Digit Dial (RDD) Sample
A list of randomly-generated telephone numbers for a given geographic area (defined by area codes, prefixes, and blocks)
Random Sample
A sample of persons selected using an independent random sampling technique to participate in a research study
Rating Scale
A means of measuring interval or ratio data, wherein respondents rate a target on a given attribute using scaled values; for example, "Please rate your satisfaction with your current physician using a 10-point scale, where 10 means you are completely satisfied and 1 means you are not at all satisfied"
Regression Analysis
A statistical technique used to measure the ability of several independent variables to predict the value of a dependent variable; for example, a regression analysis could be used to predict which website characteristics are most influential in a person's likelihood of returning to the site
Reliability
A quality of measurement that suggests that the same data would have been collected each time a study is performed
Respondent
The person participating in the research study; the research subject or participant
Response Rate
The percentage of persons who actually respond to a survey out of the total number of persons who were asked to respond
BACK TO TOP
S
Sample
A subset of the population
Sample Error
The chance that the results of a study are due to misrepresentativeness of the sample caused by random chance; for example, out of 1 million flips of a penny, the penny will land heads-up 50% of the time and tails-up 50% of the time. If we look at a sample of 100 of those 1 million flips, heads could come up 70% of the time even though, over the long-term, it will only occur 50% of the time-this difference is due to sample error.
Sample Frame
A list from which the final sample for the study is drawn; the sample frame must be representative of the population, else the sample will not be representative of the population; for example, the telephone book can be used as a sample frame, but it does not include people without a telephone, thereby biasing the sample drawn from that frame against people without telephones
Sample Size
The number of people in the sample. Sample size is a key determinant in sample error-as sample size increases, sample error decreases
Secondary Research
Research gathered from sources other than directly from the population, e.g. publications, associations, government research
Self-Response Survey
A survey that relies on the respondent to complete and return the survey alone without the assistance of an interviewer, e.g. mail surveys, online surveys
Skip Patterns
Steps in a questionnaire instructing the interviewer or respondent to skip particular questions based on their previous responses
SPSS
A statistical software package developed for use in the social sciences
Statistical Significance
A measurement of the likelihood that an observed effect will be present in the population, based on sample size, distribution of values, and size of the effect; i.e., the difference between two values from a sample is considered statistically significant if a statistical test proves that the difference is also likely to be present in the population
Stratified Sample
A sample drawn to match the distribution of a variable in the population; for example, a stratified sample of Texas residents would include the same proportion of Houston residents as there exists in the actual population of Texas
Survey Instrument
Questionnaire
BACK TO TOP
T
Telemarketing
A direct marketing tool designed to market a product or service via the telephone; often confused with a telephone research methodology, but different from market research over the telephone because market research is designed only to collect data for research purposes, not for direct marketing purposes
Tracking Research
Research designed to measure the change in results as compared to baseline research
T-Test
A statistical test to determine if two groups are significantly different from one another
Type I Error
A false positive; incorrectly accepting a positive result to be true
Type II Error
A false negative; incorrectly accepting a negative result to be true
BACK TO TOP
V
Validity
See Internal Validity and External Validity
Variable
A quantity or function that may assume a given value or set of values
Variance
The total amount of variation in values for a given variable; for example, the average daily temperature in Dallas, Texas has a larger variance over a period of a year (there is a greater amount of variation in temperatures as the seasons change) than the average daily temperature at the South Pole (although it changes slightly day-to-day, it does not change very much)
Verbatim
A direct quote from a respondent
Virtual Focus Groups
A type of focus group that is conducted via teleconference, videoconference, or over the Internet in a chat room
BACK TO TOP
W
Weighted Sample
A sample that has been multiplied by the appropriate coefficients such that the representation of a variable in the sample matches its representation in the population; a sample may be weighted based on gender such that males and females are equally represented in the sample
BACK TO TOP
|
|
 |
|
 |
|
