Glossary (See related pages) A-B design   A single-subject experimental design in which measurements are repeatedly made until stability is presumably established (baseline), after which treatment is introduced and an appropriate number of measurements are made. A-B-A design   Same as the A-B design except a second baseline is added. A-B-A-B design   Same as an A-B-A design, except that a second treatment is added. A-B-C-B design   Same as A-B-A-B design except that the second baseline phase is replaced by a modified treatment phase. Abstract   A summary of a study that describes its most important aspects, including major results and conclusions. Accessible population   The population from which the researcher can realistically select subjects for a sample, and to which the researcher is entitled to generalize findings. Achievement test   An instrument used to measure the proficiency level of individuals in given areas of knowledge or skill. Action plan   A plan to implement change as a result of an action research study. Action research   A type of research focused on a specific local problem and resulting in an action plan to address the problem. Age-equivalent score   A score that indicates the age level for which a particular performance (score) is typical. Alpha coefficient   see Cronbach alpha. Alternating-treatment design   A single-subject design for studying two or more treatments. Analysis of covariance (ANCOVA)   A statistical technique for equating groups on one or more variables when testing for statistical significance; it adjust scores on a dependent variable for initial differences on other variables, such as pretest performance or IQ. Analysis of variance (ANOVA)   A statistical technique for determining the statistical significance of differences among means; it can be used with two or more groups. Anecdotal records   Records of observed behaviors written down in the form of anecdotes. The best anecdotes tell exactly what the participant did or said without making evaluative statements in the process of reporting this information. Aptitude test   An instrument used to predict performance in a future situation. Associational research   A general type of research in which a researcher looks for relationships having predictive and/or explanatory power. Both correctional and causal-comparative studies are examples. Assumption   Any important assertion presumed to be true but not actually verified; major assumptions should be described in one of the first sections of a research proposal or report. Attitude scale   A set of statements to which the participant responds. Average   A number representing the typical score attained by a group of subjects. See measures of central tendency. B-A-B design   The same as an A-B-A-B design, except that the initial baseline phase is omitted. Background question   Question asked by an interviewer or on a questionnaire to obtain information about a respondent's background (age, occupation, etc.). Bar graph   A graphic way of illustrating differences among groups. Baseline   The graphic record of measurements taken prior to introduction of an intervention in a time-series design. Behavior questions   See experience questions. Bias   See researcher bias. Bibliography   A list of references that pertain to a topic. Biography/biographical study   A form of qualitative research in which the researcher works with the individual to clarify important life experiences Case study   A form of qualitative research in which a single individual or example is studies through extensive data collection. Categorical data/variables   Data (variables) that differ only in kind, not in amount or degree. Causal-comparative research   Research to determine the cause for, or consequences of, existing differences in groups of individuals; also referred to as ex post facto research. Census   An attempt to acquire data from each and every member of a population. Chaos theory   A theory and methodology of science that emphasizes the rarity of general laws, the need for very large data bases, and the importance of studying exceptions to overall patterns. Chi-square test   A non parametric test of statistical significance appropriate when the data are in the form of frequency counts; it compares frequencies actually observed in a study with expected frequencies to see if they are significantly different. Closed-ended question   A question and a list of alternative responses form which the respondent selects; also referred to as a closed-form item. Cluster sampling/cluster random sampling   The selection of groups of individuals, called clusters, rather than single individuals. All individuals in a cluster are included in the sample; the clusters are preferably selected randomly from the larger population of clusters. Coding   The specification of categories in content analysis research. May be done ahead of time or emerge from familiarity with the raw data. Coefficient of determination (r2)   The square of the correlation coefficient. It indicates the degree of relationship between two variables. Coefficient of multiple correlation   An index of the strength of the relationship among a combination of predictor variables and the criterion variable. Like the usual correlation coefficient, a coefficient of zero would indicate that the variables are not related. On the other extreme, a coefficient of one would indicate that scores on the criterion variable can be perfectly predicted from the set of predictor variables. Cohort study   A design (in survey research) in which a particular population is studied over time by taking different random samples at various points in time. The population remains conceptually the same, but individuals change (for example, graduates of San Francisco State University surveyed 10, 20 and 30 years after graduation). Collective case study   One that studies multiple cases at the same time. Comparison group   The group in a research study that receives a different treatment from that of the experimental group. Computer search of the literature   A method whereby key terms are used to locate research literature about a topic. Concurrent validity (evidence of)   The degree to which the scores on an instrument are related to the scores on another instrument administered at the same time, or to some other criterion available at the same time. Confidence interval   An interval used to estimate a parameter that is constructed in such a way that the interval has a predetermined probability of including the parameter. Confirming sample   In qualitative research; a sample selected to validate or extend previous findings. Constant   A characteristic that has the same value for all individuals. Constitutive definition   The explanation of the meaning of a term by using other words to describe what is meant. Construct-related validity (evidence of)   The degree to which an instrument measures an intended hypothetical psychological construct, or nonobservable trait. Content analysis   A method of studying human behavior indirectly by analyzing communications, usually through a process of categorization. Content-related validity (evidence of)   The degree to which an instrument logically appears to measure an intended variable; it is determined to expert judgment. Contextualization   Placing information/data into a larger perspective, especially in ethnography. Contingency coefficient   An index of relationship derived from a crossbreak table. Contingency question   A question whose answer depends on the answer to a prior question. Contingency table   See crossbreak table. Control group   The group in a research study that is treated "as usual." Control   Efforts on the part of the researcher to remove the effects of any variable other than the independent variable that might affect performance on a dependent variable. Convenience sample   A sample that is easily accessible. Correlation coefficient (r)   A decimal number between .00 and +1.00 and –1.00 that indicates the degree to which two quantitative variables are related. Correlational research   Research that involves collecting data in order to determine the degree to which a relationship exists between two or more variables. Counterbalanced design   A design in which all groups receive all treatments. Each group receives the treatments in a different order, and all groups are posttested after each treatment. Criterion variable   The variable that is predicted in a prediction study; also any variables used to assess the criterion-related validity of an instrument. Criterion-referenced instrument   An instrument that specifies a particular goal, or criterion, for students to achieve. Criterion-related evidence of validity (evidence of)   The degree to which performance on an instrument is related to performance on other instruments intended to measure the same variable, or to other variables logically related to the variable being measured. Critical researchers   Researchers who raise philosophical and ethical questions about the way educational research is conducted. Critical sample   In qualitative research; a sample considered to be enlightening because it is unusual. Cronback alpha   An internal consistency or reliability coefficient for an instrument requiring only one test administration. Crossbreak table   A table that shows all combinations of two or more categorical variables and portrays the relationship (if any) between the variables. Cross-sectional survey   A survey in which data are collected at one point in time from a predetermined population or populations. Cross-validation   Validation of a prediction equation with at least one group other than the group on which it was based. Crystallization   Occasions, especially in ethnography, when different kinds of data 'fall in place' to make a coherent picture. Culture   The sum of a social group's observable patterns of behavior and/or their customs, beliefs and knowledge. Curvilinear relationship   A relationship shown in a scatterplot in which the line that best fits the points is not straight. Data analysis   The process of simplifying data in order to make it comprehensible. Data   Any information obtained about a sample or a population. Degrees of freedom   A number indicating how many instances out of a given number of instance are "free to vary"—that is, not predetermined. Demographics   Characteristics of a sample or population (e.g., age, ethnicity, education). Dependent variable   A variable affected or expected to be affected by the independent variable; also called "criterion" or "outcome variable." Derived scores   A score obtained from a raw score in order to aid in interpretation. Derived scores provide a quantitative measure of each student's performance relative to a comparison group. Descriptive field notes   Notes that describe what the researcher has observed. Descriptive studies   Research to describe existing conditions without analyzing relationships among variables. Descriptors   Terms used to locate sources during a computer search of the literature. Directional hypothesis   A relational hypothesis stated in such a manner that a direction, often indicated by "greater than" or "less than," is hypothesized for the results. Discriminant function analysis   A statistical procedure for predicting group membership (a categorical variable) from two or more quantitative variables. Ecological generalizibility   The degree to which results can be generalized to environments and conditions outside the research setting. Effect size (ES)   An index used to indicate the magnitude of an obtained result or relationship. Emic perspective   The view of reality of a cultural 'insider'; especially in ethnography. Empirical   Based on observable evidence. Equivalent forms   Two tests identical in every way except for the actual items included. Equivalent-forms method   A method to obtain to reliability coefficient; a way of checking consistency by correlating scores on equivalent forms of an instrument. It is also referred to as alternate-forms reliability. Errors of measurement   Inconsistency of individual scores on the same instrument. Eta   An index that indicates the degree of a curvilinear relationship. Ethnography/ethnographic research   The collection of data on many variables over an extended period of time in a naturalistic setting, usually using observation and interviews. Etic perspective   The 'outsider' or 'objective' view of a culture's reality, especially in ethnography. Expectancy table   A table used to analyze data obtained from a categorical variable and a criterion that is categorical. Experience questions   Questions a researcher asks to find out what sorts of things an individual is doing or has done. Experiment   A research study in which one or more independent variables is systematically varied by the researcher to determine the effects of this variation. Experimental group   The group in a research study that receives the treatment (or method) of special interest in the study. Experimental research   Research in which at least one independent variable is manipulated, other relevant variables are controlled, and the effect on one of more dependent variables is observed. Experimental variable   The variable that is manipulated (systematically altered) in an intervention study by the researcher. Explanatory mixed method design   A study in which quantitative data are collected first and findings tested with subsequent quantitative data. Exploratory mixed method design   A study in which qualitative data are collected first and further clarified with qualitative data. External audit   An individual outside the study is asked to review the methods and interpretations of a qualitative study. External criticism   Evaluation of the genuineness of a document in historical research. External validity   The degree to which results are generalizable, or applicable, to groups and environments outside the research setting. Extraneous event(s)   See history threat. Extraneous variable   A variable that makes possible an alternative explanation of results; an uncontrolled variable. Extraneous variable   A variable that makes possible an alternative explanation of results; an uncontrolled variable. Factor analysis   A statistical method for reducing a set of variables to a smaller number of factors. Factorial design   An experimental design that involves two or more independent variables (at least one of which is manipulated) in order to study the effects of the variables individually, and in interaction with each other, upon a dependent variable. Feelings questions   Questions researchers ask to find out how people feel about things. Field diary   A personal statement of a researcher's opinions about people and events he or she comes in contact with during research. Field jottings   Quick notes taken by an ethnographer. Field log   A running account of how an ethnographer plans to, and actually does, spend his or her time in the field. Field notes   The notes researchers take about what they observe and think about in the field. Findings   see results (of a study). Five-number summary   Consists of the lowest score, the first quartile, the median, the third quartile, and the highest score. This summary provides a quick overview about the central tendency, variability, and shape of the distribution with just five numbers. Flowchart   Types of tally sheets used to indicate the frequency and direction of a participant's remarks. Focus group interview   An interview conducted with a group in which respondents hear the views of each other. Follow-up study   A study conducted to determine the characteristics of a group after some period of time. Foreshadowed problems   The problem or topic that serves, in a general way, as the focus for a qualitative inquiry. Frequency distribution   A tabular method of showing all the scores obtained by a group of individuals. Frequency polygon   A graphic method of showing all of the scores obtained by a group of individuals. Friedman two-way analysis of variance   A nonparametric inferential statistic used to compare two or more groups that are not independent. Gain score   The difference between the pretest and posttest scores of a measure. General references   Sources that researchers use to identify more specific references (e.g., indexes, abstracts). Generalizing   See ecological generalizibility; population generalizability. Grade-equivalent score   A score that indicates the grade level for which a particular performance (score) is typical. Grounded theory   A form of qualitative research which derives interpretations inductively from raw data with continual interplay between data and emerging interpretations. Hawthorne effect   A positive effect of an intervention resulting from the subjects' knowledge that they are involved in a study or their feeling that they are in some way receiving "special" attention. Histogram   A graphic representation, consisting of rectangles, of the scores in a distribution; the height of each rectangle indicates the frequency of each score, or group of scores. Historical research   The systematic collection and objective evaluation of data related to past occurrences to determine causes, effects, or trends of those events that may help explain present events and anticipate future events. History threat   The possibility that results are due to an event that is not part of an intervention, but which may affect performance on the dependent variable, thereby affecting internal validity. Holistic perspective   The attempt to incorporate all aspects of a culture into an ethnographic interpretation. Homogeneous sample   In qualitative research, a sample selected in which all members are similar with respect to one or more characteristics. Hypothesis   A tentative, testable assertion regarding the occurrence of certain behaviors, phenomena, or events; a prediction of study outcomes. Implementation threat   The possibility that results are due to variations in the implementation of the treatment in an intervention study, thereby affecting internal validity. Independent variable   A variable that affects (or is presumed to affect) the dependent variable under study and is included in the research design so that its effect can be determined; sometimes called the "experimental" or "treatment" variable. Inferential statistics   Data analysis techniques for determining how likely it is that results based on a sample or samples are similar to results that would have been obtained for the entire population. Informal interviews   Less-structured form of interview, usually conducted by qualitative researchers. They do not involve any specific type or sequence of questioning, but resemble more the give and take of a casual conversation. Instrument decay   Changes in instrumentation over time that may affect the internal validity of a study. Instrument   Any device for systematically collecting data, such as a test, a questionnaire, or an interview schedule. Instrumental case study   One that focuses on a particular individual or situation with little effort to generalize. Instrumentation threat   The possibility that results are due to variations in the way data are collected, thereby affecting internal validity. Instrumentation   Instruments and procedures used in collecting data in a study. Interaction   An effect created by unique combinations of two or more independent variables; systematically evaluated in a factorial design. Interjudge reliability   The consistency of two (or more) independent scorers, raters, or observers. Internal criticism   Determining if the contents of a document are accurate. Internal validity   The degree to which observed differences on the dependent variable are directly related to the independent variable, not to some other (uncontrollable) variable. Internal-consistency methods   Procedures for estimating reliability of scores using only one administration of the instrument. Interval scale   A measurement scale that, in addition to ordering scores from high to low, also establishes a uniform unit in the scale so that equal distance between two scores is of equal magnitude. Intervention study/research   A general type of research in which variables are manipulated in order to study the effect on one of more dependent variables. Intervention   A specified treatment or method that is intended to modify one or more dependent variables. Interview   A form of data collection in which individuals or groups are questioned orally. Intrinsic case study   One that attempts to generalize beyond the particular case. Item validity   The degree to which each of the items in an instrument measures the intended variable. Justification (of a study)   A rationale statement in which a researcher indicates why the study is important to conduct; includes implications for theory and/or practice. Key actors   see key informants Key informants   Individuals identified as expert sources of information, especially in qualitative research. Knowledge questions   Questions interviewers ask to find out what factual information a respondent possesses about a particular topic. Kruskal-Wallis one-way analysis of variance   A nonparametric inferential statistic used to compare two or more independent groups for statistical significance of differences. Kuder-Richardson approaches   Procedures for determining an estimate of the internal consistency reliability of a test or other instrument from a single administration of the test without splitting the test into halves. Latent content   The underlying meaning of a communication. Level of confidence   The probability associated with a confidence interval; the probability that the interval will contain the corresponding parameter. Commonly used confidence levels in educational research are the 95 and 99 percent confidence levels. Level of significance   The probability that a discrepancy between a sample statistic and a specified population parameter is due to sampling error, or chance. Commonly used significance levels in educational research are .05 and .01. Likert scale   A self-reporting instrument in which an individual responds to a series of statements by indicating the extent of agreement. Each choice is given a numerical value, and the total score is presumed to indicate the attitude or belief in question. Limitation   An aspect of a study that the researcher knows may influence the results or generalizability of the results, but over which he or she has no control. Linear relationship   A relationship in which an increase (or decrease) in one variable is associated with a corresponding increase (or decrease) in another variable. Literature review   The systematic identification, location, and analysis of documents containing information related to a research problem. Location threat   The possibility that results are due to characteristics of the setting or location in which a study is conducted, thereby producing a threat to internal validity. Logic   Using knowledge to create new knowledge. Longitudinal survey   A study in which information is collected at different points in time in order to study changes over time (usually of considerable length, such a several months or years). Manifest content   The obvious meaning of a communication. Manipulated variable   See experimental variable. Mann-Whitney U test   A nonparametric inferential statistic used to determine whether two uncorrelated groups differ significantly. Matching   Consists of two groups of items listed in columns. Respondents are required to match the item in the left column that corresponds most closely with an item in the right column. Matching design   A technique for equating groups on one or more variables, resulting in each member of one group having a direct counterpart in another group. Maturation threat   The possibility that results are due to changes that occur in subjects as a direct result of the passage of time and that may affect their performance on the dependent variable, thereby affecting internal validity. Maximal variation sample   In qualitative research, a sample selected in order to represent diversity in one or more characteristics. Mean/arithmetic mean   The sum of the scores in a distribution divided by the number of scores in the distribution; the most commonly used measure of central tendency. Measures of central tendency   Indices representing the average or typical score attained by a group of subjects; the most commonly used in educational research are the mean and the median. Measures of variability   Indices indicating how spread out the scores are in a distribution. Those most commonly used in educational research are the range, standard deviation, and variance. Mechanical matching   A process of pairing two persons whose scores on a particular variable are similar. Median   That point in a distribution having 50 percent of the scores above it and 50 percent of the scores below it. Member checking   Participants in a qualitative study are asked to check the accuracy of the research report. Meta-analysis   A statistical procedure for combining the results of several studies on the same topic. Mixed-method design   A study combining quantitative and qualitative methods. Mode   The score that occurs most frequently in a distribution of scores. Moderator variable   A variable that may or may not be controlled but has an effect on the research situation. Mortality threat   The possibility that results are due to the fact that subjects who are for whatever reason "lost" to a study may differ from those who remain so that their absence has an important effect on the results of the study. Multiple analysis of covariance (MANCOVA)   An extension of analysis of covariance that incorporates two or more dependent variables in the same analysis. Multiple correlation (R)   A numerical index describing the relationship between predicted and actual scores using multiple regression. The correlation between a criterion and the "best combination" of predictors. Multiple perspectives   The recognition and acceptance of multiple views of reality, especially in ethnography. Multiple regression   A technique using a prediction equation with two or more variables in combination to predict a criterion (y = a + b1X1 + b2X2 + b3X3…). Multiple-baseline design   A single-subject experimental design in which baseline data are collected on several behaviors for one subject, after which the treatment is applied sequentially over a period of time to each behavior one at a time until all behaviors are under treatment. Also used to collect data on different subjects with regard to a single behavior, or to assess a subject's behavior in different settings. Multiple-treatment interference   The carryover or delayed effects of prior experimental treatments when individuals receive two or more experimental treatments in succession. Natural setting   A specific place in which events and interactions among individuals typically occur. Naturalistic observation   Observation in which the observer controls or manipulates nothing, and tries not to affect the observed situation in any way. Negatively skewed distribution   A distribution in which there are more extreme scores at the lower end than at the upper, or higher, end. Nominal scale   A measurement scale that classifies elements into two or more categories, the numbers indicating that the elements are different, but not according to order or magnitude. Nondirectional hypothesis   A prediction that a relationship exists without specifying its exact nature. Nonequivalent control group design   An experimental design involving at least two groups, both of which may be pretested; one group receives the experimental treatment, and both groups are posttested. Individuals are not randomly assigned to treatments. Nonparametric technique   A test of statistical significance appropriate when the data represent an ordinal or nominal scale, or when assumptions required for parametric tests cannot be met. Nonparticipant observation   Observation in which the observer is not directly involved in the situation to be observed Nonrandom sample/sampling   The selection of a sample in which every member of the population does not have an equal chance of being selected. Norm group   The sample group used to develop norms for an instrument. Normal distribution   A theoretical "bell-shaped" distribution having a wide application to both descriptive and inferential statistics. It is known or thought to portray many human characteristics in "typical" populations. Norm-referenced instrument   An instrument that permits comparison of an individual score to the scores of a group of individuals on the same instrument. Norms   Descriptive statistics that summarize the test performance of a reference group of individuals and permit meaningful comparison of individuals to the group. Null hypothesis   A statement that any difference between obtained sample statistics and specified population parameters is due to sampling error, or "chance." Objectivity   A lack of bias or prejudice. Observational data   Data obtained through direct observation. Observer bias   The possibility that an observer does not observe objectively and accurately, thus producing invalid observations and a threat to the internal validity of a study. Observer effect   The impact of an observer's presence on the behavior observed. Observer expectations   The effect that an observer's prior information can have on observational data. One-group pretest-posttest design   A weak experimental design involving one group that is pretested, exposed to a treatment, and posttested. One-shot case study design   A weak experimental design involving one group that is exposed to a treatment and then posttested. One-tailed test   The use of only one tail of the sampling distribution of statistic – used when a directional hypothesis is stated. Open-ended question   A question giving the responder complete freedom of response. Operational definition   Defining a term by stating the actions, processes, or operations used to measure or identify examples of it. Opinion questions   Questions a researcher asks to find out what people think about a topic. Opportunistic sample   In qualitative research, a sample chosen to take advantage of conditions that arise during a study. Oral statements   Some form of oral expression. Ordinal scale   A measurement scale that ranks individuals in terms of the degree to which they possess a characteristic of interest. Outcome variable   See dependent variable. Outlier   Scores or other observations that deviate or fall considerably outside most of the other scores or observation in a distribution or pattern. Panel study   A longitudinal design (in survey research) in which the same random sample is measured at different points in time. Parameter   A numerical index describing a characteristic of a population. Parametric technique   A test of significance appropriate when the data represent an interval or ratio scale of measurement and other specific assumptions have been met. Partial correlation   A method of controlling the subject characteristics threat in correlational research by statistically holding one or more variables constant. Participant observation   Observation in which the observer actually becomes a participant in the situation to be observed. Participants   Individuals whose involvement in a study can range from providing data to initiating and designing the study. Participatory action research   Action research intended not only to address a local problem but also to empower individuals and to bring about social change. Path analysis   A type of sophisticated analysis investigating causal connections among correlated variables. Pearson r   An index of correlation appropriate when the data represent either internal or ratio scales; it takes into account each and every pair of scores and produces a coefficient between .00 and either + or – 1.00. Percentile rank   An index of relative position indicating the percentage of scores that fall at or below a given score. Performance checklist   Used to keep track of behaviors that occur. Performance test   Measures an individual's performance on a particular task. Phenomenology/phenomenological research   A form of qualitative research in which the researcher attempts to identify commonalilties in the perceptions of several individuals regarding a particular phenomenon. Pie chart   A graphic method of displaying the breakdown of data into categories. Pilot study   A small-scale study administered before conducting an actual study; its purpose is to reveal defects in the research plan. Population generalizability   The extent to which the results obtained from a sample are generalizable to a larger group. Population   The group to which the researcher would like the results of a study to be generalizable; it includes all individuals with certain specified characteristics. Portraiture   A form of qualitative research in which the researcher and the individual being portrayed work together to define meaning.