Appendix
Dictionary of Public Relations Measurement and Research
This booklet was prepared and edited by Dr. Don W. Stacks, University of Miami.
Commission on Public Relations Measurement & Evaluation Dictionary Editorial Board
Patricia Bayerlein
Gagen MacDonald
Dr. Kathryn Collins
General Motors
John Gilfeather
Roper ASW
Fraser Likely
Likely Communication
Strategies Ltd.
Marcia L. Watson
Editorial Assistant
University of Miami
Dr. Walter K. Lindenmann
PR Research & Measurement
Specialist
Dr. David Michaelson
Consultant
Dr. Tom Watson
Charles Sturt University
Dr. Donald K. Wright
University of South Alabama
Dictionary of Public Relations Measurement and Research
– A –
Algorithm—s. a step-by-step problem-solving procedure, especially an established, recursive computational procedure for solving a problem in a finite number of steps
Alpha Level (a)—s. the amount of error or chance allowed in sampling or inferential testing
Analysis of Variance (ANOVA)—s. an inferential statistical test of significance for continuous measurement dependent variables against a number of groups as independent variables
Articles—m. an output, typically printed but also found on the Internet
Attitude—m. a predisposition to act or behave toward some object; a motivating factor in public relations; composed of three dimensions: affective (emotional evaluation), cognitive (knowledge evaluation), and connotative (behavioral evaluation)
Attitude Research—m. the measuring and interpreting a full range of views, sentiments, feelings, opinions, and beliefs that segments of the public may hold toward a client or product
Attitude Scale—m. a measure that targets respondent attitudes or beliefs toward some object; typically interval-level data and requires that an arbitrary or absolute midpoint (“neutral” or “neither agree nor disagree”) be provided to the respondent; also known as Likert-type or Semantic Differential measures; s. an output measured as an interval or ratio measure
Audience—m. a specified group from within a defined public targeted for influence
– B –
Bar Graph—s. a representation of a frequency distribution by means of rectangles (or other indicators) whose widths represent class intervals and whose heights represent corresponding frequencies; see also graph
Baseline—s. an initial measurement against which all subsequent measures are compared; m. a data point established for comparison at the developmental stage of a research campaign.
Behavioral Event Interview (BEI)—m. an interview technique used to solicit evidence or examples of a specific competency or skill you possess; BEI is based on the premise that a person’s past behavior is the best predictor of their future performance
Behavioral Objective—m. an objective that specifies the expected public relations campaign or program outcomes in terms of specific behaviors; s. a measure that is actionable in that it is the behavior requested (e.g., outcome) of a target audience; see also outcome
Belief—m. a long-held evaluation of some object, usually determined on the basis of its occurrence; clusters of beliefs yield attitudes
Benchmarking (Benchmark Study)—m. a measurement technique that involves having an organization learn something about its own practices, the practices of selected others, and then compares these practices
Bivariate Analysis—s. a statistical examination of the relationship between two variables
BRAD (British Rate and Data Measure)—s. British Rate and Data measure—provides circulation and advertising costs data
– C –
Campaign (Program)—m. the planning, execution, and evaluation of a public relations plan of action aimed at solving a problem
Case Study Methodology—m. an informal research methodology that gathers data on a specific individual or company or product with the analysis focused on understanding its unique qualities; is not generalizable to other cases or populations
Categorical Data—s. measurement data that are defined by their association with groups and are expressed in terms of frequencies, percentages, and proportions; see also nominal data, ordinal data
Category—m. in content analysis the part of the system where the content (units of analysis) are placed; also referred to as “subjects” or “buckets”
Causal Relationship—m. a relationship between variables in which a change in one variable forces, produces, or brings about a change in another variable; s. the result of a significant interaction term in an analysis of variance or regression, often displayed in path analyses or sequential equation models
Census—m. collection of data from every person or object in a population
Central Tendency—s. a statistic that describes the typical or average case in the distribution of a variable; see also mean, median, mode, range, standard deviation, standardized score, variance, Z-score
Characters—m. a manifest unit of analysis used in content analysis consisting of individuals or roles (e.g., occupations, roles, race)
Chi-Square (c2)—s. an inferential statistical test of significance for categorical data (nominal or ordinal)
Circulation—s. number of copies of a publication as distributed (as opposed to read)
Closed-Ended Question—m. a question that requires participants to answer selected and predetermined responses (e.g., strongly agree, agree, neither agree nor disagree, disagree, strongly disagree)
Cluster Analysis—s. an exploratory data analysis tool which aims at sorting different objects into groups in a way that the degree of association between two objects is maximal if they belong to the same group and minimal if otherwise
Clustered Sample—m. a type of probability sample that involves first breaking the population into heterogeneous subsets (or clusters), and then selecting the potential sample at random from the individual clusters
Coefficient Alpha (a)—s. a statistical test for a measurement’s reliability for interval and ratio data; also known as Cronbach’s coefficient alpha
Cohen’s Kappa—s. an intercoder reliability measure used in content analysis when there are more than two coders; see also reliability, content analysis
Cohort Survey—m. a type of longitudinal survey in which some specific group is studied over time according to some criteria that stays the same (e.g., age = 21) while the samples may differ
Column Inches—s. total length of an article if it were all one-column measured in inches (or centimeters); determines the total “share of ink” that a company or brand has achieved
Communication—m. the process that deals with the transmission and reception of intentional messages that are a part of a natural language system (e.g., words, phrases, sentences, paragraphs)
Communication Product—m. the end result of the communication product process resulting in the production and dissemination of a brochure, media release, video news release, website, speech, and so forth; see also output, outtake
Communication(s) Audit—m. a systematic review and analysis of how effectively an organization communicates with all of its major internal and external audiences by identifying these audiences, by identifying the communication programs and their communication products utilized for each audience, by determining the effectiveness of these programs and their products, and by identifying gaps in the overall existing communication program; uses accepted research techniques and methodologies; see also formal methodology, informal methodology, case study, content analysis, survey, in-depth interview, focus group, experiment, secondary, historical, participant-observation
Communication(s) Research—m. any systematic study of the relationships and patterns that are developed when people seek to share information with each other
Community Case Study—m. an informal methodology whereby the researcher takes an in-depth look at one or several communities—subsections of communities—in which an organization has an interest by impartial, trained researchers using a mix of informal research methodologies (i.e., participant-observation, role-playing, secondary analysis, content analysis, interviewing, focus groups)
Concurrent Validity—m. a measurement device’s ability to vary directly with a measure of the same construct or indirectly with a measure of an opposite construct. It allows you to show that your test is valid by comparing it with an already valid test
Confidence Interval—s. in survey methodology based on a random sampling technique, the range of values or measurement within which a population parameter is estimated to fall (e.g., for a large population we might expect answers to a question to be within ±3% of the true population answer; if 55% responded positively, the confidence interval would be from 52% to 58%); sometimes called measurement error
Confidence Level—m. in survey methodology based on a random sampling technique, the amount of confidence we can place on our confidence interval (typically set at 95%, or 95 out of 100 cases truly representing the population under study, with no more than 5 cases out of 100 misrepresenting that population); sometimes called sampling error; s. the amount of confidence a researcher has that a finding between groups or categories is statistically significant; see also statistically significant
Construct Validity—m. a dimension of measurement; s. a statistically tested form of measurement validity that seeks to establish the dimensionality of a measure; see also validity, face validity, criterion-related validity, content validity, discriminant validity, divergent validity
Content Analysis—m. an informal research methodology (and measurement tool) that systematically tracks messages (written, spoken, broadcast) and translates them into quantifiable form via a systematic approach to defining message categories through specified units of analysis; the action of breaking down message content into predetermined components (categories) to form a judgment capable of being measured
Content Validity—m. a form of measurement validity that is based on other researchers or experts evaluations of the measurement items contained in a measure; see also validity, face validity, construct validity, criterion-related validity, discriminant validity, divergent validity
Contingency Question—m. a survey question that is to be asked only to some respondents, determined by their responses to some other questions; sometimes called a “funnel question”
Contingency Table—s. a statistical table for displaying the relationship between variables in terms of frequencies and percentages; sometimes called a “cross tabulation table” or “cross tab”
Continuous Data—s. data that are measured on a continuum, usually as interval data
Contour Plot—s. a graphical technique for representing a 3-dimensional surface by plotting constant z slices, called contours, on a 2-dimensional format—that is, given a value for z, lines are drawn for connecting the (x,y) coordinates where that z value occurs; used to answer the question “How does Z change as a function of X and Y?”
Convenience Sample—m. a nonprobability sample where the respondents or objects are chosen because of availability (e.g., “man on the street”); a type of nonprobability sample in which whoever happens to be available at a given point in time is included in the sample; sometimes called a “haphazard” or “accidental” sample
Convergent Validity—s. a type of construct validity that refers to the principle that the indicators for a given construct should be at least moderately correlated among themselves; see also coefficient alpha, validity, face validity, content validity, construct-related validity, criterion-related validity, discriminant validity, divergent validity
Correlation (r)—s. a statistical test that examines the relationships between variables (may be either categorical or continuous); see also correlation coefficient, Pearson product-moment coefficient, Spearman’s rho, r
Correlation Coefficient—s. a measure of association that describes the direction and strength of a linear relationship between two variables; usually measured at the interval or ratio data level (e.g., Pearson Product Moment Coefficient, r), but can be measured at the nominal or ordinal level (e.g., Spearman’s rho)
Cost-Effectiveness—s. an outcome that may be measured in public relations research that evaluates the relation between overall expenditure (costs) and results produced, usually the ratio of changes in costs to change in effects
Cost Per Thousand (CPM)—s. cost of advertising for each 1,000 homes reached by the media
Covariation—s. a criterion for causation whereby the dependent variable takes on different values depending on the independent variable
Criterion-Related Validity—m. a form of validity that compares one measure against others known to have specified relationships with what is being measured; the highest form of measurement validity; see also validity, face validity, content validity, discriminant validity, divergent validity
Criterion Variable—m. the variable the researchers predict to; see also dependent variable
Crossbreak Analysis—s. a categorical analysis that compares the frequency of responses in individual cells from one variable against another; see also crosstabulation, frequency, frequency table
Cross-Sectional Survey—m. a survey based on observations representing a single point in time; see also snapshot survey
Crosstabs—s. statistical tables used to array the data; allows the analyst to go beyond total data into frequencies and averages as well as to make possible overall as well as subgroup analyses (e.g., comparisons of the opinions expressed by sell-side analysts with those stated by buy-side investment professionals)
Crosstabulation—s. the result of two categorical variables in a table; see also crossbreak analysis, frequency, frequency table
Cumulative Scale (Guttman Scale/Scalogram)—m. a measurement scale that assumes that when you agree with a scale item you will also agree with items that are less extreme; see also outcome, Guttman scale, Likert scale, semantic differential scale
Cyber Image Analysis—m. the measurement of Internet content via chat rooms or discussion groups in cyberspace regarding a client or product or topic; the measurement of a client’s image everywhere on the Internet
– D –
Data—m. the observations or measurements taken when evaluating a public relations campaign or program; s. the frequencies, means, percentages used to assess a campaign or program; see also nominal data, ordinal data, interval data, ratio data
Database Mining—m. a research technique utilizing existing data; see also secondary methodology
Database—s. a collection of data arranged for ease and speed of search and retrieval
Deduction—m. a philosophical logic in which specific expectations or hypotheses are developed or derived on the basis of general principles
Delphi Technique—m. a research methodology (usually survey or interview) where the researcher tries to forecast the future based on successive waves of interviews or surveys with a panel of experts in a given field as a means of building a “consensus” of expert opinion and thought relating to particular topics or issues
Demographic Analysis—m. analysis of a population in terms of special social, political, economic, and geographic subgroups (e.g., age, sex, income level, race, educational-level, place of residence, occupation)
Demographic Data—m. data that differentiate between groups of people or things (e.g., sex, race, income)
Dependent Variable—m. the variable that is measured or collected
Depth Interview—m. an extensive, probing, open-ended, largely unstructured interview, usually conducted in person or by telephone, in which respondents are encouraged to talk freely and in great detail about given subjects; also known as an “in-depth interview”; see also in-depth interview methodology
Descriptive Research—m. a form of research that gathers information in such a way as to paint a picture of what people think or do
Descriptive Statistics—s. the reduction and simplification of the numbers representing research to ease interpreting the results
Descriptive Survey—m. a type of survey that collects in quantitative form basic opinions or facts about a specified population or sample; also known as a “public opinion poll”
Design Bias—m. research design bias is introduced when the study fails to identify the validity problems or when publicity about the research fails to incorporate the researcher’s cautions
Discriminant Validity—s. a type of validity that is determined by hypothesizing and examining differential relations between a test and measures of similar or different constructs. It is the opposite of convergent validity and is also known as divergent validity; see also convergent validity, divergent validity; m. a way of establishing if a measure is measuring what it is supposed to measure; see also validity, criterion-related validity
Divergent Validity—s. see also discriminant validity
Double-Barreled Question—m. a question that attempts to measure two things at the same time; a source of measurement error
– E –
Editorial—m. the content of a publication written by a journalist, as distinct from advertising content, which is determined by an advertiser; an article expressing the editorial policy of a publication of a matter of interest (also known as a “leader” or “leading article”); space in a publication bought by an advertiser that includes journalistic copy intended to make the reader think it originates from an independent source (also known as an “advertorial”); s. an outcome or measured variable
Environmental Scanning—m. a research technique for tracking new developments in any area or field by carrying out a systematic review of what appears in professional, trade, or government publications
Equal Appearing Interval Scale—m. a measurement scale with predefined values associated with each statement
Equivalent Advertising Value (AVE)—s. equivalent cost of buying space devoted to editorial content
Error Bar—s. a graphical data analysis technique for showing the error in the dependent variable and optionally; the independent variable in a standard x-y plot
Ethnographic Research—m. an informal research methodology that relies on the tools and techniques of cultural anthropologists and sociologists to obtain a better understanding of how individuals and groups function in their natural settings; see also participant-observation
Evaluation Research—m. a form of research that determines the relative effectiveness of a public relations campaign or program by measuring program outcomes (changes in the levels of awareness, understanding, attitudes, opinions, and/or behaviors of a targeted audience or public) against a predetermined set of objectives that initially established the level or degree of change desired
Events—s. a community affairs or sponsorship output
Experimental Methodology—m. a formal research methodology that imposes strict artificial limits or boundaries on the research in order to establish some causal relationship between variables of interest; is not generalizable to a larger population
Explanatory Research—m. a form of research that seeks to explain why people say, think, feel, and act the way they do; concerned primarily with the development of public relations theory about relationships and processes; are typically deductive
Exploratory Research—m. a form of research that seeks to establish basic attitudes, opinions, and behavior patterns or facts about a specific population or sample; are typically inductive and involve extensive probing of the population or sample or data
– F –
Face Validity—m. a form of measurement validity that is based on the researcher’s knowledge of the concept being measured; the lowest form of measurement validity; see also validity, content validity, construct validity, criterion-related validity, discriminant validity, divergent validity
Facilitator—m. an individual who leads a focus group; also known as a moderator
Factor Analysis—s. a statistical tool that allows researchers to test the dimensionality of their measures; used to assess a measure’s construct validity
Fever Graph—s. a form of line graph that measures peaks and valleys of data along a continuum that is either continuous or whose classes represent categories; see also graph
Field Study Methodology—m. a formal research methodology that imposes fewer restrictions or limits or boundaries on the research in order to test some causal relationships found in experimental research and generalize them to a larger population
Filter Question—m. a question that is used to move a respondent from one question to another; a question that is used to remove a respondent from a survey or interview; see also funnel question
Focus Group Methodology—m. an informal research methodology that uses a group approach to gain an in-depth understanding of a client, object, or product; is not generalizable to other focus groups or populations
Formal Methodology—m. a set of research methodologies that allows the researcher to generalize to a larger audience but often fails to gain in-depth understanding of the client, object, or product; a set of methodologies that follow scientific or social scientific method; a set of methodologies that are deductive in nature
Formative Evaluation—m. a method of evaluating the process by which programs occur while activities are in their early stages with the intent of improving or correcting activities
Frequency—s. a descriptive statistic that represents the number of objects being counted (e.g., number of advertisements, number of people who attend an event, number of media release pickups)
Frequency Table—s. a listing of counts and percentages in tabular form; may report a single variable or multiple variables; see also crossbreak analysis, crosstabulation
F-Test—s. an inferential test of significance associated with Analysis of Variance (ANOVA); see also Analysis of Variance
Funnel Question—m. a question used in a questionnaire or schedule that moves an interviewer or respondent from one part of a survey to another (e.g., “Are you a registered voter?” If the respondent says yes, certain questions are asked and if not, then other questions are asked); see also filter question
– G –
Goal (Objective)—m. the explicit statement of intentions that supports a communication strategy and includes an intended audience/receiver, a proposed measurable outcome (or desired level of change in that audience), and a specific timeframe for that change to occur
Grand Mean—s. a descriptive statistics which represents the mean of all sample means in a study, weighted by the number of items in each sample. The grand mean treats the individuals in the different subsets (groups) as if there were no subgroups, but only individual measures in the set. The grand mean is thus simply the mean of all of the scores; see also mean
Graph—s. a graphical representation of a variable; see also bar, pie, line, fever
Gross Rating Points (GRP)—s. measures of weight or readership or audience equivalent to audience exposure among one percent of the population; see also targeted gross rating points (TGRP)
Guttman Scale (Cumulative Scale/Scalogram)—m. a measurement scale that assumes unidimensionality and that people, when faced with a choice, will also choose items less intense than the one chosen
– H –
Histogram—s. a representation of a frequency distribution by means of rectangles whose widths represent class intervals and whose heights represent corresponding frequencies; a bar chart representing a frequency distribution; heights of the bars represent observed frequencies; see also graph
Historical Methodology—m. an informal research methodology that examines the causes and effects of past events
Holsti’s Reliability Coefficient—s. a fairly simple reliability measure used in content analysis; see also reliability, content analysis, intercoder reliability, intracoder reliability, Scott’s pi, and Krippendorf’s alpha
Hypothesis—m. an expectation about the nature of things derived from theory; a prediction of how an independent variable changes a dependent variable; formally stated as a predication (e.g., males will purchase more of X than females), but tested via the null hypothesis (e.g., males and females will not differ in their purchases of X)
Hypothesis Testing—m. determining whether the expectations that a hypothesis represents are, indeed, found in the real world
– I –
Image Research—m. a research program or campaign that systematically studies people’s perceptions toward an organization, individual, product, or service; sometimes referred to as a “reputation study”
Impressions—m. the number of people who might have had the opportunity to be exposed to a story that has appeared in the media; also known as “opportunity to see” (OTS); s. usually refers to the total audited circulation of a publication or the audience reach of a broadcast vehicle
Incidence—s. the frequency with which a condition or event occurs in a given time and population or sample
Independent t-Test—s. an inferential statistical test of significance that compares two levels of an independent variable against a continuous measured dependent variable
Independent Variable—m. the variable against which the dependent variable is tested
In-Depth Interview Methodology—m. an informal research methodology in which an individual interviews another in a one-on-one situation
Induction—m. a philosophical logic in which general principles are developed from specific observations
Inferential Research—m. statistical analyses that test if the results observed for a sample are indicative of the population; the presentation of information that allows us to make judgments whether the research results observed in a sample generalize to the population from which the sample was drawn
Inferential Statistics—s. statistical tests that allow a researcher to say within a certain degree of confidence whether variables or groups truly differ in their response to a public relations message; see analysis of variance, chi-square, bivariate correlation, correlation, Pearson product moment correlation, Spearman’s rho, regression, path analysis, sequential equation model, t-test
Informal Methodology—m. a research methodology that does not allow the researcher to generalize to a larger audience but gains in-depth understanding of the client, object, or product
Informational Objective—m. an objective that establishes what information a target audience should know or the degree of change in knowledge levels after the conclusion of a public relations campaign or program
Inputs—m. the research information and data from both internal and external sources applied in the conception, approval, and design phases of the input stage of the communication production process
Inquiry Research—m. a formal or informal research methodology that employs systematically content analysis, survey methodology, and/or interviewing techniques to study the range and types of unsolicited inquiries that an organization may receive from customers, prospective customers, or other target audience groups
Instrumental Error—m. in measurement, error that occurs because the measuring instrument was poorly written; s. tested for via reliability analyses; see also coefficient alpha, KR-20
Intercoder Reliability—m. the reliability of content analysis coding when the coding is done by two or more coders; see also reliability, intracoder reliability, Holsti’s Reliability Coefficient, Scott’s pi, Krippendorf’s alpha, Cohen’s kappa
Interval Data—m. measurement data that are defined on a continuum and assumed to have equal spacing between data points; s. includes temperature scale, standardized intelligence test scores, Likert scale, semantic differential scale, Guttman Scalogram; see also attitude research, attitude scale, data, Likert scale, Guttman scale
Interview Schedule—m. a guideline for asking questions in person or over the telephone interviewers are tasked with predicting your likelihood of success in a given position and use your past behavior as one indicator of your future performance
Intracoder reliability—m. the reliability of content analysis coding when the coding is done by only one coder, usually the researcher; s. obtained from statistical tests that analyze coder decisions versus chance; see also reliability, intercoder reliability, Cohen’s kappa, Holsti’s reliability coefficient, Krippendorf’s alpha, Scott’s pi
Issues Research—m. a formal or informal research methodology that systematically studies public policy questions of the day, with the chief focus on those public policy matters whose definition and contending positions are still evolving
Items—s. a manifest unit of analysis used in content analysis consisting of an entire message itself (e.g., an advertisement, story, press release)
– J –
Judgmental Sample—m. a type of nonprobability sample in which individuals are deliberately selected for inclusion in the sample by the researcher because they have special knowledge, position, characteristics or represent other relevant dimensions of the population that are deemed important to study; see also purposive sample
– K –
Key Performance (Performance Result)—m. the desired end effect or impact of a program of campaign performance
Known Group t-Test—s. an inferential statistical test of significance that compares the results for a sampled group on some continuous measurement dependent variable against a known value; see also inferential statistics, independent t-test
KR-20—s. a reliability statistic for nominal- or ordinal-level measurement; also known as Kuder-Richardson formula 20; see also reliability, coefficient alpha
Krippendorf’s Alpha—s. a fairly simple content analysis coding reliability measure; see also reliability, intercoder reliability, Intracoder reliability, Holsti’s Reliability Coefficient, Scott’s pi, Cohen’s kappa
– L –
Latent Content—m. from content analysis, an analysis of the underlying idea, thesis, or theme of content; the deeper meanings that are intended or perceived in a message
Likert Scale—m. an interval-level measurement scale that requires people to respond to statements on a set of predetermined reactions, usually strongly agree, agree, neither agree nor disagree, disagree, strongly disagree; must possess an odd number of reaction words or phrases; also called “summated ratings method” because the scale requires at least two, if not three, statements per measurement dimension
Line Graph—s. a representation of frequency distribution by means of lines representing data points at various intervals along a continuum; see also graph
Longitudinal Survey—m. a type of survey that consists of different individuals or objects that is observed or measured over time (e.g., multiple snapshot samples)
– M –
Mail Survey—m. a survey technique whereby a questionnaire is sent to a respondent via the mail (or Internet) and the respondent self-administers the questionnaire and then sends it back
Mall Intercept Research—m. a special type of person-to-person surveying in which in-person interviewing is conducted by approaching prospective participants as they stroll through shopping centers or malls; a nonprobability form of sampling
Manifest Content—m. from content analysis, an analysis of the actual content of a message exactly as it appears as opposed to latent content that must be inferred from messages
Market Research—m. any systematic study of buying or selling behavior
Mean—s. a descriptive statistic of central tendency that describes the “average” of a set of numbers on a continuum; also called “average”; the process of applying a precise number or metric, which is both valid and reliable, to the evaluation of some performance
Measurement—m. a way of giving an activity a precise dimension, generally by comparison to some standard; usually done in a quantifiable or numerical manner; see also data, scale
Measurement Bias—m. failure to control for the effects of data collection and measurement (e.g. tendency of people to give socially desirable answers)
Measurement Error—m. the amount of error found in a research campaign; in surveys it is the amount of error in individual responses; s. a term that expresses the amount of doubt that a researcher may accept in terms of findings; see also confidence interval
Measurement Reliability—m. the extent to which a measurement scale measures the same thing over time; s. a statistical reporting of how reliable a measure is; see also coefficient alpha, test-retest reliability, split-half reliability
Measurement Validity—m. the extent to which a measurement scale actually measures what it believed to measure; see also face validity, content validity, construct validity, criterion-related validity
Media—m. includes newspapers, business and consumer magazines and other publications, radio and television, the Internet; company reports, news wires, government reports and brochures; Internet websites and discussion groups
Media Evaluations—m. the systematic appraisal of a company’s reputation, products or services, or those of its competitors, as measured by their presence in the media
Median—s. a descriptive statistic of central tendency indicating the midpoint in a series of data; the point above and below which 50 percent of the data values fall
Mention Prominence—s. an outcome based on an indication of how prominent a company, product, or issue was mentioned in the media; typically measured in percent of article and position within the output (e.g., headline, above the fold, first three minutes)
Mentions—s. an output or outcome consisting of counts of incidents of a company or product or person appearing in the media, one mention constitutes a media placement
Message Content—m. the verbal, visual, and audio elements of a message; the material from which content analyses are conducted; s. a trend analysis factor that measures what, if any, of planned messages are actually contained in the media; see also message content analysis
Message Content Analysis—m. analysis of media coverage of messages regarding a client, product, or topic on key issues
Message Strength—s. trend analysis factor that measures how strongly a message about a client or product or topic was communicated
Mode—s. a descriptive statistic of central tendency indicating the most frequently occurring (the most typical) value in a data series
Moderator—m. an individual who leads a focus group; also known as a facilitator
Monitoring—m. a process by which data are systematically and regularly collected about a research program over time; see also environmental scanning
Motivational Objective—m. an objective that establishes the desired level of change in a target audience’s specific attitudes or beliefs after a public relations campaign
Multiple Regression—s. a statistical technique that employs multiple dependent variables to predict an outcome variable (dependent variable); see also regression, independent variable, dependent variable
Multivariate Analysis—s. an inferential or descriptive statistic that examines the relationship among three or more variables
– N –
Network Analysis—m. a formal or informal research method that examines how individuals or units or actors relate to each other in some systematic way
Neutral Point—s. a point midway between extremes in attitude measurement scales; in Likert scales usually defined as “neutral” or “neither agree nor disagree”; see also attitude, attitude scale, Likert scale, semantic differential scale
Nominal Data—s. measurement data that are simple categories in which items are different in name only and do not possess any ordering; data that are mutually exhaustive and exclusive; the simplest or lowest of all data; categorical data; example: male or female, where neither is seen as better as or larger than the other
Nonparametric Statistics—s. inferential and descriptive statistics based on categorical data; see also chi-square, Spearman’s rho
Non-Probability Sample—m. a sample drawn from a population whereby respondents or objects do not have an equal chance of being selected for observation or measurement
Nonverbal Communication—m. that aspect of the communication that deals with the transmission and reception of messages that are not a part of a natural language system (e.g., visual, spoken [as opposed to verbal], environmental)
Norm—s. short for “normative data”; see also normative data
Normal Curve—s. measurement data reflecting the hypothetical distribution of data points or cases based on interval-or ratio-level data that are “normally distributed” and error free; all continuous or parametric data sets have their own normally distributed data that fall under its specific normal curve
Normative Data—s. the proprietary set of scores that allow comparison of results to other studies and see “where you stand” and provide a context
Null Hypothesis—s. the hypothesis of no difference that is formally tested in a research campaign or program; its rejection is the test of the theory; it is the formal hypothesis that all inferential statistics test; see also inferential statistics
– O –
Objective—m. a measurable outcome in three forms: informational (cognitive), motivational (attitudinal/belief), behavioral (actionable); an explicit statement of intentions that supports a communication strategy, and to be measurable, includes an intended audience/public, a proposed change in a communication effect, a precise indication of the amount or level of change and a specific time frame for the change to occur
Omnibus Survey—m. an “all-purpose” national consumer poll usually conducted on a regular schedule (once a week or every other week) by major market research firms; also called “piggyback” or “shared-cost” survey
Open-Ended Question—m. open-ended questions probe the dimensions of attitudes and behavior held by a particular respondent through an interactive conversation between respondent and interviewer
Opinion—m. a verbalized or written evaluation of some object
Opportunities to See (OTS)—m. the number of times a particular audience has the potential to view a message, subject, or issue; s. an outcome statistic based on outputs serving as a dependent variable in some research; see also dependent variable, impressions, outcomes, output
Ordinal Data—s. measurement data that are categories in which items are different in name and possess an ordering of some sort; data that are mutually exhaustive and exclusive and ordered; categorical data; example: income as categories of under $25K, $26K–$50K, $51K–$75K, $76K–$100K, over $100K
Outcomes—m. quantifiable changes in awareness, knowledge, attitude, opinion, and behavior levels that occur as a result of a public relations program or campaign; an effect, consequence, or impact of a set or program of communication activities or products, and may be either short term (immediate) or long term; s. the dependent variable in research; see also dependent variable
Outgrowth—m. the culminating effect of all communication programs and products on the positioning of an organization in the minds of its stakeholders or publics; s. an outcome statistic used as a dependent variable in some research; see also dependent variable, outcomes
Output—m. what is generated as a result of a PR program or campaign that impacts on a target audience or public to act or behave in some way—this is deemed important to the researcher (also known as a “judgmental sample”); the final stage of a communication product, production, or process resulting in the production and dissemination of a communication product (brochure, media release, website, speech, etc.); s. the number of communication products or services resulting from a communication production process; the number distributed and/or the number reaching a targeted audience; sometimes used as an outcome serving as a dependent variable in research; see also dependent variable, outcome
Outtake—m. measurement of what audiences have understood and/or heeded and/or responded to a communication product’s call to seek further information from PR messages prior to measuring an outcome; audience reaction to the receipt of a communication product, including favorability of the product, recall and retention of the message embedded in the product, and whether the audience heeded or responded to a call for information or action within the message; s. sometimes used as an outcome serving as a dependent variable in research; see also dependent variable, outcome
– P –
Paired t-Test—s. an inferential statistical test of significance that compares data that are collected twice on the same sample; see also inferential statistics, independent t-test, known-group t-test
Panel Survey—m. a type of survey that consists of the same individuals or objects that are observed or measured over time; a type of survey in which a group of individuals are deliberately recruited by a research firm because of their special demographic characteristics for the express purpose of being interviewed more than once over a period of time for various clients on a broad array of different topics or subjects
Parameter—s. in sampling, a characteristic of a population that is of interest
Parametric Statistics—s. inferential and descriptive statistics based on continuous data; see also data, descriptive statistics, inferential statistics
Participant-Observation—m. an informal research methodology where the researcher takes an active role in the life of an organization or community, observes and records interactions, and then analyzes those interactions
Path Analysis—s. a statistical technique that establishes relationships between variables with arrows between variables indicating the pattern of causal relationships usually in the form of a “path diagram”; see also path diagram
Path Diagram—s. a graphical representation of the causal relationships between variables showing both direction and strength of relationship
Pearson Product-Moment Coefficient (r)—s. a correlation statistic used with interval and ratio data; see also correlation, data, Spearman’s rho
Percentage—s. a descriptive statistic based on categorical data; defined as the frequency count for a particular category divided by the total frequency count; example: 10 males out of 100 people = 10%; see also descriptive statistics
Percentage Point—s. the number that a percentage is increased or decreased
Percent of Change—s. a measure of increase or decrease of media coverage
Performance—m. the act of carrying out, doing, executing, or putting into effect; a deed, task, action, or activity as a unit of a program of performance
Performance Indicator—m. a sign or parameter that, if tracked over time, provides information about the ongoing results of a particular program of performance or campaign; s. an outcome measured during a public relations campaign that serves as a dependent variable; see also data, dependent variable
Performance Measure—m. a number that shows the exact extent to which a result was achieved; s. in a research campaign, an outcome of some sort serving as a dependent variable; see also data, dependent variable, outcomes
Performance Result (Key Performance)—m. the desired end effect or impact of a program of campaign performance
Performance Target—m. a time-bounded and measurable commitment toward achieving a desired result
Periodicity—s. a bias found in sampling due to the way in which the items or respondents are chosen; example: newspapers may differ by being daily, weekly, weekday only, and so forth
Pie Graph—s. a representation of a frequency distribution by means of triangular portions of a pie whose sections represents the percentages of the variable of interest; see also graph
Piggyback Survey—m. see omnibus survey
Poll—m. a form of survey research that focuses more on immediate behavior than attitudes; a very short survey-like method whose questionnaire asks only very short and closed-ended questions; see also in-depth interview survey, survey methodology
Positioning—m. trend analysis factor that measures how a client or product or topic was positioned in the media (e.g., leader, follower)
Position Papers—m. print output
Probability Sample—m. a sample drawn at random from a population such that all possible respondents or objects have an equal chance of being selected for observation or measurement
Probe Question—m. a question used in a questionnaire or schedule that requires the participant to explain an earlier response, often in the form of “why do you think this?”
Product (Communication Product)—m. the end result of the communication product or process resulting in the production and dissemination of a brochure, media release, video news release, website, speech, and so forth; an output or outtake; see also output, outtake
Program (Campaign)—m. the planning, execution, and evaluation of a public relations plan of action aimed at solving a problem
Prominence of Mention—m. trend analysis factor that measures how prominently a client or product or topic was mentioned and where that mention occurred (e.g., headline, top of the fold, what part of a broadcast); s. an output unit of analysis used as a dependent variable; see also dependent variable, output
Proportion—s. a descriptive statistic based on categorical data; defined as the percentage as made part of one (1.0); example: 10 males out of 100 people are 10 hundredths of the sample
PR Return on Investment (PRROI)—m. the impact of a public relations program on business investment; s. the outcome (dependent) variable that demonstrates the impact of a public relations campaign or program investment on the overall business outcomes; a causal indicator of public relations impact; see also causal relationships, Return on Investment (ROI)
Psychographic Research—m. research focusing on a population or sample’s nondemographic traits and characteristics, such as personality type, lifestyle, social roles, values, attitudes, and beliefs
Psychometrics—s. a branch of psychology that deals with the design, administration, and interpretation of quantitative tests for the measurement of psychological variables such as intelligence, aptitude, and personality traits; also called psychometry
Public—m. a group of people who have consequences for an organization or are affected by the consequences of organizational decisions; a group of people from which the public relations campaign or program selects specific targeted audiences in an attempt to influence it regarding a company, product, issue, or individual; see also audience, sample
Public Opinion Poll—m. a type of survey that collects basic opinions or facts about a specified population or sample; also known as a descriptive survey; see also poll, survey methodology
Public Relations Effectiveness—s. the degree to which the outcome of a public relations program is consonant with the overall objectives of the program as judged by some measure of causation; see also causal relationship.
Purposive Sample—m. a nonprobability sample in which individuals are deliberately selected for inclusion based on their special knowledge, position, characteristics, or relevant dimensions of the population
Push Poll—m. a survey technique in which an interviewer begins by acting as if the telephone call is a general survey but then asks the respondent a question implying questionable behaviors or outcomes of a person or product
– Q –
Q-Sort—m. a measurement instrument that focuses on respondents’ beliefs by asking them to sort through piles of opinion statement and sort them into piles on an 11-point continuum usually bounded by “most-like-me” to “most-unlike-me”; see also attitude scale
Qualitative Research—m. usually refers to studies that are somewhat to totally subjective, but nevertheless in-depth, using a probing, open-ended, response format or reflects an ethnomethodological orientation
Quantitative Research—m. usually refers to studies that are highly objective and projectable, using closed-ended, forced-choice questionnaires; research that relies heavily on statistics and numerical measures
Question—m. a statement or phrase used in a questionnaire or schedule that elicits either an open- or closed-ended response from a research participant; see also funnel questions, probe questions
Questionnaire—m. a measurement instrument that contains exact questions and measures an interviewer or survey researcher uses to survey through the mail, Internet, in person, or via the telephone; may be closed-ended and open-ended, but typically employs more closed- ended questions
Quota Sample—m. a type of nonprobability sample that draws its sample based on a percentage or quota from the population and stops sampling when that quota is met; a nonprobability sample that attempts to have the same general distribution of population characteristics as in the sample; see also poll, survey methodology
– R –
Range—s. a descriptive central tendency statistic that expresses the difference between the highest and lowest scores in the data set; example: responses to a question on a 1 to 5 Likert scale where all reaction categories were used would yield a range of 4 (5 minus 1)
Ratio Data—s. measurement data that are defined on a continuum and possess an absolute zero point; examples: number of children, a bank account, absolute lack of heat (0° Kelvin = –459.67° or –273.15° C)
Reach—m. refers to the scope or range of distribution and thus coverage that a given communication product has in a targeted audience group; broadcasting, the net unduplicated (also called “duplicated”) radio or TV audience for programs or commercials as measured for a specific time period
Readership—m. number of people who actually read each issue of a publication on average; s. an outcome variable that often serves as a dependent variable; see also dependent variable, outcome
Regression—s. a statistical tool that predicts outcomes based on one outcome (dependent) variable and one predictor (independent) variable; see also multiple regression; m. a source of error or invalidity in experimental methodology that may impact on the validity of the experiment; see also experimental methodology, validity, inferential statistics
Reliability—m. the extent to which results would be consistent, or replicable, if the research were conducted a number of times; s. a statistical measure accessing consistency of a measure, usually through the coefficient alpha or KR-20 statistic in measurement or Cohen’s kappa, Hosti’s reliability coefficient, Krippendorf’s alpha, or Scott’s pi; see also measurement reliability, Cohen’s Kappa, Holsti’s reliability coefficient, Scott’s pi
Reputation—s. an outcome variable often used dependent variable in public relations research dealing with the public’s perception of some source’s credibility, trustworthiness, or image based on the source’s behavior; see also dependent variable
Research Bias—m. unknown or unacknowledged error created during the design, measurement, sampling, procedure, or choice of problem studied; see also experimental methodology, validity, regression
Research Instrument—m. tool used to collect data; see also questionnaire, interview schedule, semistructured interview, structured interview
Research—m. the systematic effort before (formative research) or during and/or after (summative or evaluative research) a communication activity aimed at discovering and collecting the facts or opinions pertaining to an identified issue, need, or question; may be formal or informal
Respondent—m. the individual from whom data are collected through participation in a research campaign; sometimes called participant or, in psychological study, subject
Response Rate—m. from survey methodology, the number of respondents who actually completed an interview; s. the percentage of completed surveys (often adjusted for mailing errors)
Results—s. the outcome demonstrated to have been impacted upon by a public relations campaign; m. that which is measured in a campaign as dependent variables; see also dependent variable, outcome, output, outtake, outgrowth
Return on Investment (ROI)—s. an outcome variable that equates profit from investment; see also public relations effectiveness, dependent variable
– S –
Sample—m. a group of people or objects chosen from a larger population; see also probability sample, nonprobability sample; convenience sample; panel survey; longitudinal survey; snapshot survey
Sampling Error—m. the amount of error expected or observed in surveys that may be attributed to problems in selecting respondents; s. the amount of error that is acceptable or expected based on the sample size and expressed as confidence in sampling form a population; see also confidence level
Scale—m. a measurement instrument consisting of attitude or belief items that reflect an underlying structure toward some attitude or belief object; see also attitude scale
Scalogram (Guttman Scale/Cumulative Scale)—m. a measurement scale that assumes (a) unidimensionality and (b) that people, when faced with a choice, will also choose items less intense than the one chosen; see also attitude scale, Likert scale, semantic differential scale
Scattergram—s. a descriptive statistics based on continuous data that graphically demonstrated how data are distributed between two variables; also known as a scatter diagram or scatterplot
Schedule—m. the timeline on which a public relations program or campaign is conducted; a list of questions, usually open-ended, used in focus group and in-depth interviews to gather data; see also survey methodology, in-depth interview methodology
Scott’s pi—s. a coding reliability measure employed in content analysis that reduces the impact of chance agreement among intercoder or intracoder coding; see also reliability, content analysis, Holsti’s reliability coefficient, Krippendorf’s alpha, Cohen’s kappa
Screener Question—m. one of several questions usually asked at the beginning of an interview or survey to determine if the potential respondent is eligible to participate in the study; see also funnel question
Secondary Methodology—m. an informal research methodology that examines extant data in order to draw conclusions; a systematic reanalysis of a vast array of existing data; often used in benchmarking and benchmark studies
Semantic Differential Scale—m. an attitude measure that asks respondents to evaluate an attitude object based on bipolar adjectives or phrases separated by a continuum represented as consisting of an odd number of intervals; developed by Osgood, Suci, and Tannenbaum; see also attitude scale, Guttman scale, Likert scale
Semantic Space—m. the idea that people can evaluate attitude objects along some spatial continuum; often associated with attitude researchers Osgood, Suci, and Tannenbaum
Semistructured Interview—m. an interview conducted with a fairly open framework that allows for focused, conversational, two-way communication; it can be used both to give and receive information
Sequential Equation Model—s. a statistical methodology similar to path analysis but that uses as measures that are created such as attitude, intelligence, reputation rather than actual indicators (e.g., sales, revenue) to test a hypothesized causal relationship between predictor (independent) and outcome (dependent) variables; see also dependent variable, independent variable, path analysis, regression, multiple regression
Share of Ink (SOI)—s. measurement of the total press/magazine coverage found in articles or mentions devoted to a particular industry or topic as analyzed to determine what percent of outputs or Opportunities to See (OTS) is devoted to a client or product; an outcome often used as a dependent variable; see also dependent variable, outcome
Share of Voice (SOV)—s. measurement of total coverage devoted to radio/television coverage to a particular industry or topic as analyzed to determine what percent of outputs or Opportunities to See (OTS) is devoted to a client or product; also known as “share of coverage”; an outcome often used as a dependent variable; see also dependent variable, outcome
Shared-Cost Survey—m. see omnibus survey
Simple Random Sample—m. a type of probability sample in which numbers are assigned to each member of a population, a random set of numbers is generated, and then only those members having the random numbers are included in the sample
Situation Analysis—m. an impartial, often third-party assessment of the public relations and/or public affairs problems, or opportunities, that an organization may be facing at a given point in time
Skip Interval—m. the distance between people selected from a population based on systematic sampling; usually defined as the total population divided by the number of people to be sampled (e.g., for a sample of 100 people to be drawn from a population of 10,000 people, the skip interval would be 100/10,000 = 100 individuals skipped between selected participants)
Snapshot Survey—m. a type of survey that consists of individuals or objects that is observed or measured once; see also cross-sectional survey
Snowball Sample—m. a type of nonprobability sample in which individuals who are interviewed are asked to suggest other individuals for further interviewing
Sociogram—s. a pictorial representation of the actual relationships of individuals within a specified unit such as a public, target audience, or work unit
Sources Mentioned—m. trend analysis factor that measures who was quoted in media coverage; also known as “quoteds”
Speaking Engagements—s. print or broadcast or Internet communication product output; see also output
Spearman’s rho—s. a correlation statistic used with nominal or ordinal data; see also correlation, data, Pearson product-moment coefficient
Split-Half Reliability—s. a test for a measure’s reliability where a sample is randomly split and one segment receives a part of the measure and the second segment receives the rest
Standard Deviation (SD)—s. a descriptive statistic of central tendency that indexes the variability of a distribution; the range from the mean within which approximately 34% of the cases fall, provided the values are distributed in a normal curve
Standardized Score (Z-Score)—s. a descriptive statistic based on continuous data that expresses individual scores based on their standard deviations from the group mean; range of scores is usually –3.00 to +3.00; see also z-score
Statistical Significance—s. refers to the degree to which relationships observed in a sample can be attributed to sampling error or measurement error alone; expressed in terms of confidence that the relationships are due to error X% of the time (e.g., 5%); expressed in terms of the confidence that we have that the results are due to what was measured X% of the time (e.g., 95% confident); see also inferential statistics, confidence interval
Stratified Sample—m. a type of probability sample that involves first breaking the total population into homogenous subsets (or strata) and then selecting the potential sample at random from the individual strata; example: stratifying on race would require breaking the population into racial strata and then randomly sampling within each stratum
Structured Interview—m. an interview with a predefined set of questions and responses that may provide more reliable, quantifiable data than an open-ended interview and can be designed rigorously to avoid biases in the line of questioning
Summary Measure—s. summary measures combine information of different types and from different sources, which together, permit a rapid appraisal of a specific phenomenon to identify differences (e.g. between groups, countries), observed changes over time, or expected changes (e.g. as a consequence of policy measures); there are four key elements to summary measures: the selection of relevant parameters to be included, the reliable measurement/collection of these parameters, the unit in which the summary measure will be expressed, and the relative weight of each of the constituents in the total summary measure
Summative Evaluation—m. a method of evaluating the end of a research program; the basis of establishing the dependent measures; see also dependent variable
Survey Methodology—m. a formal research methodology that seeks to gather data and analyze a population’s or sample’s attitudes, beliefs, and opinions; data are gathered in person or by telephone (face-to-face), or self-administered via mail, e-mail, or fax; see also survey methodology, longitudinal survey, panel survey, cohort survey, snapshot survey
Symbols/Words—s. a manifest unit of analysis used in content analysis consisting of specific words (e.g., pronouns, client name, logotypes) that are counted; see also content analysis
Systematic Sample—m. a type of probability sample in which units in a population are selected from an available list at a fixed interval after a random start
– T –
Target Audience—m. a very specific audience differentiated from “audience” by some measurable characteristic or attribute (e.g., sports fishermen)
Targeted Gross Rating Points (TGRP)—s. gross rating points (GRP) targeted to a particular group or target audience; an outcome often used as a dependent variable; see also dependent variable, gross rating points, outcome
Test-Retest Reliability—s. a test for a measure’s reliability by testing the same sample with the same measure over time
Themes—s. a latent unit of analysis used in content analysis that measures an underlying theme or thesis (e.g., sexuality, violence, credibility); see also content analysis
Throughputs—m. the development, creative, and production activities (writing, editing, creative design, printing, fabrication, etc.) as part of a throughput stage of a communication product production process
Time/Space Measures—s. a manifest unit of analysis used in content analysis consisting of physically measurable units (e.g., column inches, size of photographs, broadcast time for a story); see also content analysis
Tone—s. trend and latent content analysis factor that measures how a target audience feels about the client or product or topic; typically defined as positive, neutral/balanced, or negative; often used as an outcome and dependent variable; see also dependent variable, outcome, content analysis
Trend Analysis—m. tracking of performance over the course of a PR campaign or program; survey method whereby a topic or subject is examined over a period of time through repeated surveys of independently selected samples (snapshot or cross-sectional survey)
t-Test—s. an inferential statistical test of significance for continuous measurement dependent variables against a bivariate independent variable; used when total number of observations are less than 100; see also paired t-test; independent t-test; known group t-test, inferential statistics
Type of Article—m. categories of a publication such as “product review,” “by-lined article,” “editorial,” “advertorial,” and “feature story”; s. trend analysis factor that measures the nature of client or product or topic coverage (e.g., column inches, broadcast time); often used as a dependent variable; see also dependent variable
– U –
Unit of Analysis—m. the specification of what is to be counted in content analysis methodology; consist of symbols/words, time/space measures, characters, themes, and items; may be manifest (observable) or latent (attitudinal)
Univariate Analysis—s. the examination of only one variable at a time
Universe—m. the set of all the units from which a sample is drawn; also called the population
– V –
Validity—m. the extent to which a research project actually measures what it is intended, or purports to measure; see also measurement validity
Value—m. an underlying cultural expectation that usually directs an individual’s beliefs
Variance (s2)—s. a descriptive statistic of central tendency that measures the extent to which individual scores in a data set differ from each other; the sum of the squared standard deviations from the mean
Verbatim—m. a transcript of the actual comments participants make in a focus group or individuals. Many researchers include verbatims in their final reports to support their interpretation of the finding; s. data that may be used in content analysis; see also interview schedule, semistructured interview, structured interview, content analysis
– W –
Weighted Average—s. an average that takes into account the proportional relevance of each component rather than treating each component equally
Weighting—s. assigning a numerical coefficient to an item to express its relative importance in a frequency distribution
Word/Symbol—s. from content analysis, a unit of analysis consisting of the actual word or symbol communicated in the media; see also content analysis
– Z –
Z-Score (Standardized Score)—s. a descriptive statistic of central tendency that takes data from different types of scales and standardizes them as areas under the normal curve for comparison purposes; see also standardized score
The Commission on Public Relations Measurement & Evaluation
Formed under the auspices of the Institute for Public Relations, the Commission exists to establish standards and methods for public relations research and measurement, and to issue authoritative best-practices white papers.
The Institute for Public Relations
This independent foundation is dedicated to the science beneath the art of public relations. The Institute exists to build and document research-based knowledge in the field of public relations, and to mainstream this knowledge by making it available and useful to practitioners, educators, researchers, and the clients they serve.
Patricia Bayerlein
Gagen MacDonald
Dr. Kathryn Collins
General Motors
John Gilfeather
Roper ASW
Fraser Likely
Likely Communication Strategies Ltd.
Marcia L. Watson Editorial Assistant
University of Miami
Dr. Walter K. Lindenmann
PR Research & Measurement Specialist
Dr. David Michaelson
Consultant
Dr. Tom Watson
Charles Sturt University
Dr. Donald K. Wright
University of South Alabama
A wide array of papers authored by members of the Commission on Public Relations Measurement & Evaluation can be found at http://www.instituteforpr.org.