-
case study is the presentation of data about a single setting or event. It is not a method of research as such because the data being offered can have been gathered using a variety of different methods (questionnaire, observation, and so forth). It is predominantly a description, and is usually based on a qualitative data set, though statistics, such as survey findings, may be incorporated.
-
A count and record of how many people live in a certain area. A census often includes information such as address, age, birth date, sex, etc. for each person.
-
A type of rater effect in which an assessor or an evaluator tends to rate teachers toward the mid-point of a scale or to judge the performance as average or neutral when it is actually well above or well below the middle level of the scale. This use of term central tendency is not the same as that used in statistics.
-
A statistical test that measures significance in the study of frequency distributions.
-
A study that puts people or things into a small number of separate groups, so that there will be as much likeness within each group, and as much difference among the groups as possible.
-
To select naturally occurring groups within a population (e.g. classes within a school).
-
Coded data refers to a way of recording material at data collection, either manually or on computer, for analysis. The data is put into groups or categories, such as age groups, and each category is given a code number. Data is usually coded for convenience, speed, computer storage space and to permit statistical analysis.
-
A study of a group of people who stay in that group over a long period of time. For example, all people born in 1960 are a cohort, as are all students who will graduate from high school in 1999. The study follows this group over time, rather than looking at them once.
-
The state of having more than one disease, diagnosis, or disorder at the same time; for example, the same person might have depression and panic attacks during the same period.
-
The measure of how well the test being studied and the "gold standard" test measure the same thing, at the same time.
-
A number range that shows how likely it is that the true amount is inside the listed range of amounts; for example, a 95% confidence interval of 25-45 would mean there is a 95% chance that the right amount (number, score, measurement) is somewhere between 25 and 45.
-
A promise from the interviewer to the person being interviewed that no information will be given to anyone except the researchers, if it can show who the person being interviewed is. For example, an interviewer could promise an ex-patient that complaints the ex-patient makes about a doctor would never reach anyone not working on the study, unless the complaint were completely separated from the ex-patient who made it. There are almost always some limits to confidentiality; for example, if the person being interviewed is thought to be a“danger to self or others,” then that information could be given after all. If an interviewer thinks a person is suicidal, the interviewer could tell a doctor so, even if the interview is confidential.
-
The inability to tell between the separate impacts of two or more factors on a single outcome. For example, one may find it difficult to tell between the separate impacts of genetics and environmental factors on depression.
-
A general idea that tries to explain something; for example, social status is a construct.
-
The measure of how well the test fits the ideas behind the study and the way the topic has been set out. Usually such a test separates two groups that are known to be opposite extremes.
-
A kind of study that picks out themes by noticing the details of books, newspapers, movies, speeches, etc.; for example, counting the number of times a word or phrase is used in President Clinton’s speeches.
-
The measure of how fully the whole topic of the study is covered by the test . For a test to have content validity, every piece of the topic should also be part of the test. This is sometimes called “face validity.”
-
Something that has an unlimited number of possible values; for example, height, weight, and age are all continuous because a person’s height, weight, or age could be measured in smaller and smaller fractions between the numbers of the whole inches, pounds, or years.
-
This is a feature of experimental research and is there to provide a contrast to the experimental group through the removal of the independent variable. The use of a control group may be necessary in order to measure the validity of a research finding.
-
Researchers may control some variables in order to allow the research to focus on specific variables, without being distorted by the impact of the excluded variables. A common way to control a variable is to be selective; for example, gender is controlled by selecting as respondents only men or only women; age can be partially controlled by restricting a sample to one age range, rather than any age.
-
The measure of how well the test matches up with other tests of the same thing.
-
A measure, ranging from 0.00 to 1.00, of how well two or more things (“variables,” values, scores, etc.) change together. Both things may get higher at the same time, or lower at the same time, or one may get higher while the other gets lower. For example, saving money and spending money are correlated (inversely), because the more money you save, the less you spend.
-
When changes in one variable are accompanied by changes in another variable. For example, if a person takes a vocabulary test and a reading comprehension test, changes in scores on one test might be accompanied by changes in scores on the other test.
-
The measure of how well the test matches an accepted test (a “gold standard”) outside the study.
-
A number that shows whether all the items on a scale or test are related and pulling in the same direction.
-
The degree in which similarities and differences in the characteristics of participants from different groups can be assessed.
-
The administration of two or more experimental therapies, one after the other, in a specified or random order to the same group of people.
-
Studies in which participants are evaluated over short periods of time.
-
Cross-tabulating is the process of analysing data according to one or more key variables. A common example is to analyse data by the gender of the research subject or respondent, so that you can compare findings for men with findings for women. Also known as cross-referencing.
-
A method used to prove the validity of a test by administering it a second time on a new selected group from the same population.
-
Some experimental designs are said to be "crossed"; that is, each level of each factor appears with each level of all others. For example, if you were trying to test two types of drugs on two types of virus, each type of drug would be used with each type of virus.
-
Skills that help researchers to understand and appreciate cultural differences among different groups. Cultural competence requires that researchers draw on values and customs within the community they are studying, and that they work with people of and from that community.
-
A book that lists the American Psychiatric Association definitions of mental disorders (illnesses); usually coordinated with the International Classification System.
-
In everyday conversation "data" and "information" are often used as meaning much the same thing, but sometimes they are used differently. In research it is common to refer to the raw material gathered as "data", and it is then processed manually or on computer. It becomes "information" when it acquires meaning through aggregation or interpretation by the researcher or automated analysis. Some depict a hierarchy of: data --> information --> knowledge.
-
Data analysis is subjecting data to a systematic analysis which can range from statistical to textual analysis, either manually or electronically.
-
The number of values or amounts that are free to vary in one calculation. Degrees of freedom are used in the formulas that test hypotheses statistically.
-
Information about the sample that includes areas such as age, sex, social class, presence of children, etc.
-
The “effect” that depends on changes in the “cause” (or “independent variable”). In an experiment, the dependent variable is the one the researcher measures. For example, better sleep might be dependent and a change in medication would be independent.
-
Describes certain characteristics of populations, and identifies and explores relationships between variables.
-
A way of sharing information by putting numbers into words so the information is easier to understand.
-
Something that makes something else change. For example, what you eat can make you have more red blood cells, and if red blood cells were being studied, the food you ate would be a determinant.
-
Developmental research is a variant of applied research in that the research has a problem solving function and leads to further research on the basis of its own findings.
-
A kind of measurement in which possible answers or things are ordered from lowest to highest. For example, if possible answers ranged from “1 — completely agree” to “10 — completely disagree,” someone taking the survey could choose any number from 1 to 10.
-
Separate values or groupings, with no possible values (numbers, measurements) between them. The only choices are separate categories; for example, “male” and “female” are discrete variables.
-
The mechanisms by which the results of research are communicated to stakeholders and other interested parties.
-
Any questions or events which divert attention from what is being tested; usually items in a questionnaire to keep subjects from understanding what is being tested.
-
The measure of how often something (for example, an age or a hair color) is found in the group being studied; or the range of those measures.
-
Two different questions asked as if they were only one question; for example, “Did you take this medication and was it helpful?” which should be: “Did you take this medication?” (first question) and “Was it helpful?” (second question).
-
Two different diagnoses given to the same person. This term is used most often for a “severe mental illness" combined with some form of “chemical dependency” (for example, schizophrenia combined with dependence on marijuana).