Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
149 Cards in this Set
- Front
- Back
|
Action research
|
Research conducted by practitioners in applied fields (teaching, counseling, etc.) for the purpose of solving a specific problem.
|
|
|
Affective measures
|
Educational tests that measure noncognitive constructs such as traits, attitudes, and beliefs.
|
|
|
Agreement index
|
A measure of the extent to which raters' ratings match in an absolute sense.
|
|
|
Alternative hypothesis
|
A rival explanation for the research results (often resulting from lack of control or procedural influences). It is an explanation "alternative" to that suggested by the research hypothesis and often "explains away" the findings.
|
|
|
Artifact
|
An unwanted source of influence creating a plausible, rival explanation of results. An artifact introduces biases and distorts results.
|
|
|
Attribute variable
|
A measured characteristic of research participants (e.g. learning style) presumed to be related to a dependent variable and part of the research hypothesis (If A, then B qualified by C, where C refers to the attribute variable).
|
|
|
Bivariate design
|
A type of correlational research design that assesses relationships using two measures.
|
|
|
Blinding
|
A control procedure that reduces bias by assuring that data collectors and/or research participants do not have information that distorts perceptions or influences behavior.
|
|
|
Case study design
|
An approach to qualitative research that focuses on the study of a single person or entity using an extensive variety of data.
|
|
|
Casual comparative research
|
Nonintervention research aimed at uncovering relationships by comparing groups of people who already differ on a variable of interest. It uses designs that search for causes or effects of a preexisting factor of interests.
|
|
|
Central tendency
|
A characteristic of a distribution of scores that describes where scores tend to center. It is often referred to as an average the most common one being the mean.
|
|
|
Chi-square test
|
A statistical test that determines significant relationships using data in the form of frequencies (counts, tallies, etc.)
|
|
|
Cognitive measures
|
Educational tests that measure mental abilities and school achievement.
|
|
|
Confounding
|
The "co-mingling" of an independent and extraneous variable such that as the levels of an independent variable change, so do differences in an extraneous variable (e.g., 10-,20-, 30-minute lectures given in cold, comfortable, and hot classrooms, respectively).
|
|
|
Construct
|
A label for an abstract trait or ability (such as creativity or intelligence) that is only presumed to exist, as it cannot be measured directly.
|
|
|
Construct validity
|
The extent to which a test reflect the trait or ability that the test is presumed to measure.
|
|
|
Content Validity
|
The extent to which a test reflects the domain of content that it presumably samples.
|
|
|
Control
|
The idea that procedures used in research can minimize bias, neutralize threats to validity, rule out alternative explanations, and help establish cause-and-effect relationships.
|
|
|
Control group
|
A group not receiving treatment, one that functions as a comparison so that a treatment effect can be isolated from extraneous influences.
|
|
|
Correlation coefficient
|
A statistical index of the degree of linear association between two measures (ranging from -1.00 to +1.00) revealing its strength and direction
|
|
|
Correlational research
|
A type of non-experimental research using one of several designs that measure individual differences in an attempt to uncover relationships between variables.
|
|
|
Counterbalanced quasi-experiment
|
A type of quasi-experimental research design that establishes control by using a single group to test all treatment and control conditions. It is also called repeated measures or within subject design.
|
|
|
Counterbalancing
|
A control procedure that assures order and carryover effects are neutralized by arranging conditions to be equally affected by order and carryover influences.
|
|
|
Credibility
|
A criterion for judging the trustworthiness of qualitative data and analysis.
|
|
|
Critical thinking
|
Careful, astute reasoning that includes conceptualization, synthesis, analysis, and evaluation. Critical thinkers reach sound conclusions about the connections between ideas and observations.
|
|
|
Cross-sectional survey design
|
A research design that describes changes over time using respondents who reflect temporal differences such as age (but data collection occurs at one point in time).
|
|
|
Deductive reasoning
|
A chain of logic progressing from general (e.g. theory) to specific (e.g. hypothesis) that is often used in quantitative research.
|
|
|
Degree of freedom
|
A mathematical concept used in statistics that conveys information about size (sample size, number of groups, etc.).
|
|
|
Dependent variable
|
A variable reflecting the presumed effect of the manipulation of an independent variable (e.g., score on an achievement test). It is the B in the If A, then B hypothesis.
|
|
|
Descriptive research
|
Research aimed at describing the characteristics of a population without generalizing or testing statistical hypotheses.
|
|
|
Discrepant findings
|
Research outcomes that are inconsistent (or in opposition) across studies. Such findings are often explained by differences in research methods and procedures.
|
|
|
Dispersion
|
A characteristic of a distribution of scores that describes the spread or scatter of scores around the central tendancy. The most common measure of dispersion is the standard deviation. The higher standard deviation, the greater the spread of scores around the mean.
|
|
|
Ecological generalization
|
The extent to which research findings extend beyond the settings which produced the sample data.
|
|
|
Effect size (d)
|
An index of a treatment effect expressed as a standardized difference between two means (the mean difference divided by the standard deviation of a comparison group's mean). It is often expressed as a percentile shift or "boost" from a baseline at the 50th percentile.
|
|
|
Emergent methodology
|
An approach to qualitative data analysis that relies on inductive reasoning and continual interplay between data and developing interpretation.
|
|
|
ERIC
|
Educational Resources Information Center, a database of journal and non-journal education literature and documents.
|
|
|
Error score
|
A theoretical construct referring to the difference between a true score and an observed score.
|
|
|
Ethnographic design
|
A nonexperimental descriptive research design that is usually participatory and extends over a period of time in a natural setting. It often uses observational methods, interviews, and a variety of other qualitative methods.
|
|
|
Evaluation research
|
Research concerned with assessing the value of a program or revising it to enhance its effectiveness.
|
|
|
Experimental research
|
Research involving an independent variable - a manipulation of some type (a treatment or intervention).
|
|
|
Experimeter expectancy
|
Bias that influences researchers in ways that create conditions favoring expected findings.
|
|
|
External validity
|
The degree to which research results can be generalized beyond the sample and conditions that yielded the findings.
|
|
|
Extraneous events
|
A threat to internal validity that includes influences co-occurring with a treatment between a pretest and posttest.
|
|
|
Extraneous variable
|
Any variable external to a research situation that is impacting the research and requires control so that its influence is neutralized.
|
|
|
F test
|
A statistical test also known as the analysis of variance (ANOVA) that determines the statistical significance of two or more mean differences.
|
|
|
Fourfold table
|
A method of displaying data to reveal a pattern between two variables, each with two categories of variation.
|
|
|
Frequency distribution
|
A plot of scores displayed by their frequency of occurrence (i.e. ranked).
|
|
|
Grounded theory
|
An approach to qualitative data analysis using emergent methodologies, such as constant comparison, that permits a theory to develop from the data (from the ground up) without preconceived or inflexible ideas.
|
|
|
Group research
|
Research aimed at studying one or more large groups to learn more about relationships among variables or trends over time.
|
|
|
Hawthorne effect
|
Bias that influences research participants' behavior stemming from a treatment's unintended effects related to special attention, novelty, or similar treatment co-occurrences.
|
|
|
Historical research
|
The collection and objective study of documents and artifacts related to a past event, often includingn a description of patterns or trends, in an attempt to explain a phenomenon or test a hypothesis with present-day relevance.
|
|
|
Independent variable
|
A variable reflecting a presumed cause and the conditions created (manipulated) by the researcher for comparison (e.g., type of teaching method). It is the A in the If A, then B hypothesis.
|
|
|
Inductive reasoning
|
A chain of logic progressing from specific (e.g., observations) to general (e.g., theory) that is often used in qualitative research.
|
|
|
Inference
|
A form of logic used in statistics that permits a conclusion about a population based on data collected from a sample.
|
|
|
Inferential research
|
Research aimed at generalizing to a larger population with data collected from samples of the population.
|
|
|
Inferential statistics
|
Statistical reasoning that permits generalization beyond the sample to a larger population. Central to this reasoning is the notion of statistical significance, meaning that a relationship found in the sample is probably not due to the workings of chance.
|
|
|
Instrumentation
|
A threat to internal validity that includes changes in the measuring device or measuring procedures between a pretest and posttest. It also refers to the process of gathering data with the use of measuring tools such as tests or surveys.
|
|
|
Interaction effect
|
An effect that occurs in factorial designs when the influence of one factor depends on the level or category of a second factor (e.g., a treatment affects males but not females). The keyword here is depends.
|
|
|
Internal consistency reliability
|
Test consistency determined by a correlating test items within a single test (e.g., coefficient alpha).
|
|
|
Internal validity
|
The degree to which conditions and procedures establish control. These conditions and procedures rule out rival hypotheses, reduce bias, and neutralize unwanted, potentially contaminating influences. A study has internal validity to the extent that the outcome can be explained by variation introduced by the treatment (and not an uncontrolled variable).
|
|
|
Inter-rater reliability
|
Raters relative consistency determined by correlating two or more raters' ratings.
|
|
|
Intuition
|
A belief without an empirical basis. Research findings often contradict intuitive beliefs.
|
|
|
John Henry effect
|
Bias that influences control or comparison groups' performance due to a perceived threat or similar negative perception.
|
|
|
Large-scale policy research
|
Research using large data sets (often standardized measures) with the intention of shaping public policy and influencing reform.
|
|
|
Level of an independent variable
|
The catagories or conditions that define an independent variable (e.g., 10-,20-, and 30-minute lectures if the independent variable is lecture length).
|
|
|
Longitudinal survey design
|
A descriptive research design that collects data from the same group of respondents over a period of time. Such designs often focus on the process of change.
|
|
|
Main effect
|
An effect that occurs in factorial designs when the averages between categories of one factor are different overall (e.g., females averaged over treatment and control groups score differently than males averaged over treatment and control groups). Each factor in a factorial design permits an evaluation of its main effect. The keyword here is overall.
|
|
|
Margin of error
|
A statistical index used in survey results to convey an interval that likely includes true population value.
|
|
|
Matched comparison group design
|
A type of quasi-experimental research design that arranges for group similarity.
|
|
|
Matching
|
A control procedure that equates groups (or materials, etc.) by assuring they are the same or similar on important variables.
|
|
|
Meaningful research
|
Research conducted with clear value, which is often assessed in terms of positive impact in areas such as social and educational influence. It may also change ways of thinking or suggest new directions for research and theory.
|
|
|
Meta-analysis
|
A statistical summary of many studies conducted to answer the same research questions.
|
|
|
Mortality
|
A threat to internal validity that refers to a loss of research participants between a pretest and posttest that stems from an influence of the treatment itself. Also referred to as attrition.
|
|
|
Multiple regression
|
A statistical technique using partial correlation as a control that attempts to predict an outcome (a criterion) given two or more predictor variables.
|
|
|
Negative correlation
|
A tendency of high scores on one variable to be linked to low scores on another variable (and of low scores to be linked with high scores).
|
|
|
Nonexperimental research
|
Research using designs that do not involve an intervention or a manipulation.
|
|
|
Nonparametric tests
|
Statistical tests and indexes (e.g., chi-square test, Spearman rank order correlation) that make few assumptions about the value of parameters.
|
|
|
Normal (bell) distribution
|
A common shape of score distributions. Normal distributions have common properties, ones that divide a distribution's area in terms of proportions contained within 1,2, and 3 standard deviations above and below the mean.
|
|
|
Null hypothesis
|
A statistical hypothesis asserting there is no relationship among variables being studied in the population.
|
|
|
Observational definition
|
Researcher-recorded data based on personal perceptions of research participants or objects.
|
|
|
Operational definition
|
An empirical measure of the presumed indicator of a construct of state, such as heart rate as an indicator of "anxiety."
|
|
|
P (p value)
|
An abbreviation for probability, p refers to the likelihood that chance factors, or statistical error, could explain a relationship found in a sample.
|
|
|
Parameter
|
A characteristic of a population.
|
|
|
Partial correlation
|
A control procedure that statistically equates participants on important variables to rule out the variables' confounding influences.
|
|
|
Placebo
|
A control condition that preserves the illusion of participants' receiving a treatment.
|
|
|
Population
|
A large well-defined group that generates a sample (often randomly). Data from the sample are used to make inferences about the larger population.
|
|
|
Population generalization
|
The extent to which research findings extend beyond the sample of research participants who provided data.
|
|
|
Positive correlation
|
A tendency of high scores on one variable to be linked to high scores on another variable (and of low scores to be linked with low scores).
|
|
|
Power
|
A statistical concept used in a sample size determination that refers to the likelihood of finding a significant relationship between variables in the sample resuming there is a true relationship in the population being studied.
|
|
|
Practical research
|
Research blending aspects of theory-based and problem-based research (as in a test of a "practical theory" such as multiple intelligences).
|
|
|
Predictive validity
|
The extent to which test scores accurately predict an outcome (a criterion).
|
|
|
Pre-experimental design
|
A weak research design involving a treatment but no control features (e.g., a one-group pretest-posttest design).
|
|
|
Problem-based research
|
Research focusing on direct application of findings to solve practical problems. It is also referred to as applied research.
|
|
|
Qualitative research
|
Research aimed at explaining complex phenomena through verbal descriptions rather than testing hypotheses with numerical values.
|
|
|
Quantitative research
|
Research aimed at testing hypotheses with numerical values.
|
|
|
Quasi-experimental design
|
A research design that incorporates a quasi-independent variable (an independent variable manipulation without random assignment).
|
|
|
Quasi-experimental research
|
Research involving the use of a manipulated independent variable (an intervention) without random assignment of participants to groups, weakening the researchers' ability to ferret out cause-and-effect relationships. n
|
|
|
Quasi-independent variable
|
An independent variable that does not permit random assignment of research participants.
|
|
|
Random assignment
|
Assignment of research participants to groups such that all members have an equal and independent chance of being assigned to each group.
|
|
|
Random process
|
A control procedure that assures that extraneous influences are neutralized by unsystematic scattering across people and conditions.
|
|
|
Random selection
|
A method of sampling which ensures that each member of a population has an equal and independent chance of being selected for inclusion in a sample. Variants of random selection exist, such as cluster, multistage, and stratified.
|
|
|
Randomized factorial design
|
A true experimental design that incorporates two or more factors (e.g., an independent variable and an attribute variable), permitting the testing of main effect and interaction effects.
|
|
|
Reflective teaching
|
Teachers thinking critically about the art and science of teaching, collecting data to test ideas, and revising their practice to solve classroom problems or improve learning.
|
|
|
Regression
|
A threat to internal validity that refers to a tendency of those with extreme scores to score somewhat closer to the mean upon retesting.
|
|
|
Relationship
|
Any connection between variables-though not necessarily cause and effect- whereby values of one variable ten to co-occur with values of another variable.
|
|
|
Reliability
|
The consistency of measures indexed by an estimate of errors contained within a set of scores.
|
|
|
Replicate and extend
|
A model of research favoring the repetition of an earlier study with an additional feature that enhances its generalization.
|
|
|
Research analysis
|
The "unpacking" and "dissection" of a published research study for the purpose of laying groundwork for critical review.
|
|
|
Research bias
|
Distortion of data collected in a research study that is explained by unwanted influences stemming from observers, research participants, procedures and settings, or researchers themselves.
|
|
|
Research critique
|
The descriptive review and evaluation of published research.
|
|
|
Research ethics
|
Established guidlines that encourage responsible research practices and assure the protection of human research participants.
|
|
|
Research hypothesis
|
A predicted outcome based on theory or understanding, which is often stated as If A, then B. It may also be stated as a foreshadowed question in qualitative research.
|
|
|
Research problem
|
The state of affairs or context for research that makes clear why data are being collected.
|
|
|
Research question
|
The purpose of research stated in the form of a question.
|
|
|
Researchable
|
A research hypothesis (or question) that can be tested (or answered) directly by the collection and analysis of data.
|
|
|
Review of literature
|
A summary of past and current research and thinking on a specific researchable topic.
|
|
|
Scale of measurement
|
Distinguishing features that identify adjacent values on a measuring device (includes nominal, ordinal, interval, and ratio scales).
|
|
|
Scattergram
|
A plot of paired scores revealing a visual display of a correlation. It is also known as a scatterplot or scatter diagram.
|
|
|
Selection
|
A threat to internal validity that includes groups of participants who are not comparable before the introduction of treatment.
|
|
|
Sequence effects
|
Unwanted influences due to ordering (first, second, etc.) and carryover (preceding versus following) of treatments, materials, and so on.
|
|
|
Single-subject design
|
A type of quasi-experimental research design using one subject to test all treatment and control conditions over time (e.g., an ABAB design).
|
|
|
Single-subject research
|
Research aimed at studying a single individual (or very small group) to learn more about relationships among variables or trends over time.
|
|
|
Small-scale evaluation research
|
Research aimed at evaluating local programs or procedures for the purpose of improvement or decision making.
|
|
|
Spurious relationship
|
A relationship that can be "explained away" by reference to another variable. When a connection between A and B is due solely to both being related casually to C, it is said that the relationship between A and B is spurious (or "false").
|
|
|
Standard error of mean differences
|
A measure of sampling error contained within a mean difference. conceptually, it is the standard deviation of a hypothetical sampling distribution of many mean differences drawn from a population.
|
|
|
Standard error of measurement
|
A statistical index that estimates the amount of error in a single score.
|
|
|
Statistic
|
A characteristic of a sample. Sample statistics are often used to estimate population parameters.
|
|
|
Statistical power
|
A statistical concept referring to the likelihood of correctly rejecting the null hypothesis (concluding there is a relationship in the population when, indeed, one exists).
|
|
|
Statistical significance
|
A concept referring to a finding that cannot easily be explained by the workings of chance factors. It is usually triggered when the p drops below .05.
|
|
|
Statistical test
|
A statistical maneuver applied to sampled data that yields p and permits a decision about the null hypothesis (i.e., whether to accept or reject it).
|
|
|
Strong inference
|
A concept referring to the test of a research hypothesis against an alternative (rival) hypothesis. Strong inference designs permit the elimination of at least one competing explanation (while supporting another).
|
|
|
T test
|
A common statistical test that determines whether a mean difference is statistically significant.
|
|
|
Tails
|
A statistical concept referring to the direction of an anticipated outcome (e.g., less than, more than). Most statistical tests are "two-tailed," permitting an outcome that could fall either way (like an increase or decrease). It also refers to the outer 5% limits in a normal distribution.
|
|
|
Teacher research
|
Self-reflective inquiry whereby teachers study their own practice, collect data, and attempt to solve a problem or improve learning in their classrooms.
|
|
|
Test for r
|
A statistical test that determines whether a correlation coefficient is significantly different from zero.
|
|
|
Test-retest reliability
|
Test consistency determined by correlating test scores and retest scores using the same test (stability reliability) or similar test (equivalence reliability).
|
|
|
Theory
|
An elaborated explanation for a construct or phenomenon. Theories organize empirical findings or suggest future research.
|
|
|
Theory-based research
|
Research aimed at testing the hypothesis spun from a theory with the intention of evaluating the theory (supporting or discarding it) or revising its tenets.
|
|
|
Theory-based design
|
A type of quasi-experimental research design that attempts to establish control via multiple observations of one group before and after treatment.
|
|
|
Traditional research
|
Formal scientific research using accepted guidelines and an integrated process aimed at testing hypotheses.
|
|
|
Triangulation
|
A method used in qualitative research that involves cross-checking multiple data collection sources and procedures.
|
|
|
True experimental research
|
Research involving the use of a manipulated independent variable (an intervention) coupled with random assignment of subjects to groups. Such designs are strong for testing cause-and-effect relationships.
|
|
|
True independent variable
|
An independent variable that permits random assignment of research participants.
|
|
|
True score
|
A theoretical construct referring to a person's score containing no error. It is also defined as one person's average on many tests of the same subject.
|
|
|
Type I error
|
A statistical fluke in sampling that permits an incorrect rejection of the null hypothesis (concluding there is a relationship in the population when there is no relationship).
|
|
|
Type II error
|
A statistical fluke in sampling that permits an incorrect acceptance of the null hypothesis (concluding there is no relationship in the population when in fact there is a relationship).
|
|
|
Validity
|
The meaningfulness of scores, which is often assessed by the accuracy of inferences made on the basis of test scores. It also refers to the extent to which a test measures what it is supposed to measure.
|
|
|
Variable
|
Any dimension with two or more changing values, such as age or sex.
|
|
|
Variance
|
A measure of dispersion among a set of scores (the square of the standard deviation).
|
