LIS 504 - Terms with definitions


|A |B |C |D |E |F |G |H |I |J |K |L |M |N |O |P |Q |R |S |T |U |V |W |X |Y |Z

A

acceptance region

set of values of a test statistic that imply acceptance of the null hypothesis.

action research

research aiming to contribute both to people's immediate practical concerns and to scientific knowledge by collaboration. (example)

alternative hypothesis

hypothesis to be accepted if the null hypothesis is rejected. (example)

anonymity

an ethical requirement of some research involving human subjects that the participants remain anonymous throughout the study, even to the researchers themselves. (example)

association

between two variables, relationship between variation in one and variation in the other. (example)

availability sampling

sampling in which items are selected on the basis of convenience. (example)

average

any of several measures of central tendency, most commonly the arithmetic mean, but sometimes the median. (example)

B

bias

systematic error. Types include nonresponse bias and selective observation. (example)

bibliometrics

quantitative research on documents and their relationships with one another and with other entities such as authors. (example)

C

case study

collection and presentation of detailed information about a particular individual or small group. (example)

causation

(also called causality) a relationship that, among other things, is (1) close enough to be useful or interesting (2) does not require too much qualification (3) has been shown to be not obviously spurious and (4) can be explained by a mechanism deduced from theory. To show a causal relationship, one must satisfy the time order criterion, covariation of cause and effect, and the absence of other plausible explanations. (example)

central tendency

the center of a set of measurements. Measures of central tendency include the mean the median, and the mode.

chi-square

a statistic used to test a null hypothesis that departures from the expected number of items in a set of categories arose by chance. (example)

closed question

(also called structured question) a question in which a limited number of choices are provided as allowable answers. (example)

cluster sampling

sampling by dividing the population into clusters, typically based on geography, and taking a sample of the clusters. (example)

comparison group

control group that receives a different treatment from the experimental group.

complete observer

in participant observation, a researcher who does not participate in group activities but is identified as a researcher.

concept

mental image that summarizes a set of similar observations, feelings, ideas, etc. (example)

conceptualization

process of specifying what is meant by a term.

conclusion

section of an empirical research report that interprets the results in broader terms and may suggest implications.

confidence interval

a statistic consisting of a range and a confidence level. For example, "the poll shows that 50% of the people, plus or minus 5%, will vote "yes" in the referendum; this value is accurate 19 times out of 20." (example)

confidence level

a number used to indicate how sure a researcher is that a statement about a population is correct given data about a sample.

confidentiality

an ethical principle of research involving human subjects requiring that subjects be assured of restrictions on the availability of identifying information. (example)

construct validity

a kind of validity, extent to which a measure fits with other measures into a framework of theory.

content analysis

a method involving classification and counting of elements in texts. (example)

content validity

See face validity.

contingency question

See filter question.

control group

a group that receives a standard treatment or no treatment at all and provides a base against which to measure an experimental group. (example)

correlation

See association.

covert participant

a researcher in field research whose identity as a researcher is not revealed to those studied.

critical value

a value of a test statistic that separates the rejection region and the acceptance region in a statistical test.

cross population generalizability

generalizability from findings about one population to another. (example)

cross-sectional study

a study that takes place at a single point in time.

D

data

numbers, character strings, images, etc., which can be received, stored, processed, and transmitted by humans or computers.

deductive reasoning

(also called deduction) a method, using logic, of obtaining knowledge that is implicit in other knowledge.

definition

an expression of what a word or phrase means. Types include by synonyms, by naming properties (ontological, nominal, or constitutive definition), by denotation (with examples), and operational definition. (example)

dependent variable

a variable whose change or different states the researcher wants to understand, explain, or predict. (example)

descriptive statistics

statistics which describe or summarize data, such as measures of central tendency and dispersion.

dispersion

(also called variation) extent to which values for a measure are spread out.

distribution

a list of all the categories used and the number of items in each category. (example)

double blind

research design in which neither the researchers nor the people being studied know enough to create bias.

E

ecological fallacy

a fallacy involving making conclusions about individuals based only on analyses of group data. (example)

empirical research

research involving observations and propositions derived primarily from sense experience by means of inductive reasoning. (example)

empirical research report

a formal report of empirical research, usually including the following components: introduction, literature survey, methodology, results, and conclusion.

erosion measure

measure that reflects wearing away or removal of material.

error

See also bias; noise.

error of estimation

distance between an estimate and the true value.

ethical

conforming to accepted standards of social or professional behavior. Ethical principles of research involving human subjects include voluntary participation and (usually) informed consent and confidentiality or the stricter standard of anonymity.

ethnography

long term, immersive investigation of a group. (example)

evaluation research

research with the purpose of providing useful feedback about some object.

ex post facto hypothesizing

a fallacy in which the same data used to develop a hypothesis are used to test it. (example)

exception fallacy

a fallacy involving reaching a conclusion about a group on the basis of exceptional cases.

exhaustivity

a desirable property of choices in a closed question requiring that the choices provided cover all possibilities.

experiment

research method in which the researcher intentionally manipulates one or more of the independent variables and then observes the changes in the dependent variables (example)

experimental design

the scheme by which the variables are related to one another, typically so as to achieve the greatest efficiency. (example)

experimental group

a group which receives a treatment in an experiment. (example)

expert opinion

judgments and estimates made by people who have spent much of their time working with a particular subject and filtering and storing information on it.

explanation

a statement that makes one understand an association. (example)

external validity

generalizability of results. (example)

F

face validity

(also called content validity) whether "on its face" a particular measure seems to represent the intended concept. (example)

fallacy

a mistake in logic. Types include ex post facto hypothesizing, the exception fallacy, the ecological fallacy, the hidden factor fallacy and the regression fallacy. (example)

fence sitters

respondents who choose a middle or neutral response.

field research

research in which the researcher goes "into the field" to observe the phenomenon in its natural state.

filter question

(also called a contingency question) a question asked in order to determine if the respondent is qualified to answer a subsequent question or questions. (example)

floaters

respondents who would choose a "don't know" response to a closed question if it were offered.

focus group

group interview in which discussion is encouraged. (example)

G

generalizability

(also called transferability) extent to which results of one research study can be extended to other situations. (example)

graph

a graphic representation of data. Types include the histogram and the pie chart.

grounded theory research

research involving continuous interplay between data collection and analysis.

H

Hawthorne effect

a reactive effect caused by awareness by the people being studied that they are being studied, similar to the placebo effect and a possible source of bias.

hidden factor fallacy

(also called spurious causal relationship or third variable problem) a fallacy involving erroneously concluding that x causes y when in fact a third factor z causes both x and y.

histogram

a bar graph, in which the bars are contiguous. (example)

historical research

research that records and explains past events involving human beings. (example)

hypothesis

statement that attempts to explain or predict a single phenomenon. (example)

I

idiographic

referring to laws or rules that relate to specific individuals. (example)

independent variable

a variable whose effect upon the dependent variable one is trying to understand. (example)

indicator

operation that yields the value of a variable; for example, a question on a questionnaire. (example)

inductive reasoning

a form of reasoning that moves from specific observations to broader generalizations and theories. (example)

informed consent

an ethical principle of research involving human subjects stating that subjects must be fully informed about the procedures and risks involved in research before they can consent to participate. (example)

intensive interviewing

relatively unstructured interview method seeking in-depth information. (example)

internal validity

a form of validity having to do with whether an observed association is causal in nature. (example)

interval scale

a scale in which the intervals between successive categories are significant, normally being considered as equal.

interview

data gathering method in which a researcher asks questions of participants either in person or via a medium that allows for interaction. (example)

introduction

section of a document, such as an empirical research report, that states the questions or problems under investigation and justifies their importance.

L

law

a empirically verified statement predicting an association given a set of conditions, usually fairly general. (example)

lie factor

the size of an effect shown in a graph divided by the actual size of the effect in the data on which the graph is based.

literature survey

a summary or evaluation of related research, as in a research proposal or an empirical research report.

longitudinal study

a study that takes place over time, with at least two waves of measurement. (example)

M

matching

assigning subjects to groups so that they represent the characteristics deemed relevant to as nearly equal a degree as possible.

mean

an average computed by applying a formula to a set of numeric values, usually by adding them and dividing by the number of values (the arithmetic mean). (example)

measurement validity

extent to which a measure measures what it is supposed to measure. (example)

median

an average, equal to the numeric value in a set that is bigger than half the remaining values and smaller than the other half. If there is no such single value, the mean of the two values that are bigger than half the remaining values and smaller than the other half is used instead. (example)

methodology

section of an empirical research report that describes the procedures used.

mode

a measure of central tendency equal to the value that occurs most often in a set. If there is a tie, the distribution is said to be bimodal or multimodal. (example)

multi-stage sampling

combination of sampling methods in which subunits are sampled within units. (example)

mutual exclusivity

a property of a set of classes whereby no item belongs in more than one of the classes. It is often a desirable property of the set of choices in a closed question.

N

noise

unsystematic error.

nominal scale

a scale in which the order of the categories is not significant. (example)

nomothetic

referring to laws or rules that pertain to the general case.

nonresponse bias

bias arising because some portion of a population is not available for study. (example)

normal distribution

(also called a Gaussian distribution) a distribution, bell-shaped and symmetrical about its mean, whose shape is determined by its standard deviation and the normal distribution formula*.

null hypothesis

the hypothesis under test in a statistical test. (example)

O

operational definition

a definition that gives the steps one must go through in order to observe the concept being defined. (example)

ordinal scale

a scale in which the order of categories is significant, but differences between successive categories are not. (example)

P

panel study

a study in which the same people are observed or questioned at several different times.

participant observation

(also called participant observer method) field research in which the researcher establishes a sustained relationship with people while they go about their normal activities.

percentile

The pth percentile of a set of values is the value such that p%, and no more, of the values are lower.

pie chart

(also called a circle chart) a type of chart or graph in which values are represented as segments of a circle. (example)

population

(also called universe) the set of items of interest, from which a sample may be drawn. (example)

positivism

view that objective reality exists apart from human perceptions.

posttest measure

a measure taken after application of a treatment.

predictive validity

a kind of validity, extent to which a measure can predict some other characteristic. (example)

pretest

a trial run of the methodology of a study to find out whether it works or what needs to be modified. (example)

pretest measure

a measure taken before application of a treatment. (example)

primacy effect

a sequence effect due to the fact that the first stimulus is often remembered better than the middle stimuli.

primary sampling units

units sampled at the top level in multi-stage sampling. (example)

primary source

in historical research, a document produced directly as a result of historical events, such as a diary or birth certificate, rather than indirectly from other documents.

purposive sampling

non-random sampling in which items are selected for a purpose. (example)

Q

qualitative research

empirical research using non-quantitative data, such as texts. Research in which the usual methods are qualitative includes action research, case study, ethnography, and grounded theory research. (example)

quantitive research

empirical research involving calculations based on quantitative data. (example)

quartile

The lower quartile is the 25th percentile and the upper quartile is the 75th.

questionnaire

research tool containing one or more questions to be answered by respondents. (example)

quota sampling

a type of nonrandom sampling in which interviewers are instructed to obtain interviews with fixed numbers of people in each of a number of categories, more closely specified than in stratified sampling.

R

random assignment

(also called randomization) a procedure that assigns each unit to a group based on a random event. (example)

random error

see noise

random sample

(also called simple random sample) a sample for which every combination of elements of a given size had an equal chance of being selected from a population. (example)

range

a measure of dispersion equal to the difference between the largest and the smallest value in a set.

ratio scale

a scale in which the ratios between the numbers represented by categories are significant. (example)

reactive effect

change in behavior of those being studied due to their being studied. Types include the Hawthorne effect. (example)

recency effect

a sequence effect due to the fact that the most recent stimulus is often remembered better than the middle stimuli.

referee

a subject expert to whom a manuscript submitted for publication is sent for review.

regression fallacy

a fallacy involving selecting extreme cases at one point in time, noting that the same or related cases at another point of time have values closer to the mean, and concluding incorrectly that the population as a whole is "regressing toward the mean".

rejection region

set of values of a test statistic that indicate rejection of the null hypothesis.

reliability

extent to which a measure is repeatable or stable or free from noise. Types include intra-observer reliability, inter-rater or inter-observer reliability, test-retest reliability, parallel-forms reliability, and internal consistency reliability. (example)

replication

attempt to come up with similar results under similar circumstances, success of which may be an argument in favor of generalizability. (example)

representativeness

extent to which a sample accurately reflects proportions in the population. (example)

response rate

proportion of questionnaires actually completed and returned. (example)

research

careful, systematic study and investigation in a field of knowledge. Scholarly or scientific research also involves communication with, and openness to criticism from, a broad community of researchers.

results

section of an empirical research report that summarizes the data or observations, including any computation or analysis.

S

sampling

selection of a set of items from a population as representative of the population as a whole. Types include availability sampling, cluster sampling, quota sampling, random sampling, sequential sampling, snowball sampling, stratified sampling, systematic sampling (example)

sampling error

noise due to calculating statistics from a sample rather than from the entire population.

sampling frame

empirical representation of the population of interest. (example)

secondary sampling units

units sampled directly within the primary sampling units in multi-stage sampling. (example)

selective observation

a type of bias favoring observations that confirm the researcher's pre-existing beliefs.

sequence effect

an effect arising because one observation affects a later observation. Types include the primacy effect, the recency effect, and the serial effect. (example)

sequential sampling

sampling that stops when enough data have been collected. (example)

serial effect

a sequence effect in which questions lead responses by establishing a certain tone.

snowball sampling

sampling by asking individuals studied to provide references to other potential subjects. (example)

standard deviation

a measure of dispersion, equal to the positive square root of the variance. (example)

statistical inference

probabilistic method of discounting noise as an explanation for a sample result.

stratified sampling

sampling by dividing a population into groups on the basis of some characteristic and then sampling each group.

structured question

See closed question.

survey

A data gathering method that asks one or more standard questions of a group of respondents either by interview or by questionnaire or by a combination. (example)

systematic error

See bias.

systematic sampling

sampling by choosing every nth item from a list, beginning at a random point. (example)

T

t test

(also called Student's t test) a test of significance of the means of small samples. (example)

test statistic

a value derived from the sample measurements used as a decision-maker in a test of a hypothesis. (example)

theoretical sampling

non-random sampling in which items are selected sequentially on the basis of earlier observations.

theory

organized system of accepted knowledge that applies in a variety of circumstances to explain a specific set of phenomena. (example)

time order criterion

criterion for causation that the cause must precede the effect.

transferability

See generalizability.

treatment

manipulation of one or more independent variables in an experiment.

triangulation

the use of multiple measures and methodologies to increase the likely validity of the results. (example)

type-I error

rejecting the null hypothesis when it is true.

type-II error

accepting the null hypothesis when it is false.

U

unit of analysis

the major entity type being studied, about which data are gathered; for example, individual, household, group, book edition, copy of book (example)

universe

see population.

unobtrusive measure

a measure that does not require the researcher to intrude in the research context.

V

validity

extent to which research actually studies what the researcher wants it to. Aspects and indicators of overall validity include measurement validity, predictive validity, external validity, internal validity, face validity, and construct validity.

variable

a measure in which the researcher is interested and that is expected to have different values in the different parts of the research. Main types are the independent variable and the dependent variable.

variance

a measure of dispersion equal to the arithmetic mean of the squares of the differences between the values and their arithmetic mean. It is also equal to the square of the standard deviation.

variation

See dispersion.

voluntary participation

an ethical principle of research involving human subjects forbidding coercion. (example)
Home

Last updated October 1, 2001.
This page maintained by Prof. Tim Craven
E-mail (text/plain only): t.craven@uwo.ca
Faculty of Information and Media Studies
University of Western Ontario,
London, Ontario
Canada, N6A 5B7