Skip to main content

Research Methods Simplified

Definitions

-Bayesian Methodology--can be used to measure equivalence between two rating scales, inferences of comparison (equivalent, inferior or superior) between treatments, diagnostics, etc. The methodology is grounded in subjective probability rather than relative frequency probability (Pawlowicz, 2012 p. 81)

-Bias--systematic errors that move the scores in a given direction and may result in "finding" the results the researcher wanted.

-Blind technique or masked--participant OR researcher is unaware of what conditions or treatment the participant is receiving.

-Case Study--in-depth analysis and systematic description of one patient or group of similar patients/populations used to promote an understanding of interventions.

-Construct--a mental state that cannot be directly observed or manipulated with current technology, e.g., assertiveness, intelligence, hunger, love.

-Control group--Participants randomly assigned to not receive the experimental treatment (the independent variable).  The control group is compared to the treatment group to determine the effectiveness of the treatment. 

-Cross-sectional research--data are collected for a set of cases (individuals or aggregates, e.g., cities, counties, countries) on a set of variables. Data is collected at a single time and for a single time point or interval of time.

-Dependent variable (effect)--measure of the effect of the independent variable. The independent variable is controlled and its effect is seen by the change in the dependent variable, e,g., number of years in school (independent) and reading proficiency (dependent)...think cause & effect.

-Double-blind technique--Neither the participants nor the people who have a direct contact with the participants know the type of treatment the participants have received. Used to improve validity.

-Empirical--describes observations or research based on actual observations (experiments). The idea that the only valid form of knowledge is based on the senses. Emphasis is placed on data collection. 

-Experimental group--participants who are randomly assigned to receive the treatment or event.

-Grounded theory--is a qualitative research approach used to assess how people progress or move through life experiences, i.e., life processes expressed in terms of stages or phases of adjustment, e.g., coping or adjusting to life changes such as aging, illness, new situations. "The theory remains connected to ("grounded in") the data in which it was generated through examples and explanations that show the fit between the theory and supporting empirical evidence." (Powers & Knapp, 2006).

-Hawthorne effect-- subjects become aware that they are participating in a study or experiment and this affects their behavior and thus the results of the study. (Johnson, 1995) 

-Hypothesis--any proposition advanced for testing or appraisal as a generalization about a phenomenon. (Jary, 1991)

-Independent variable (cause)--variable that is experimentally manipulated, i.e., controlled, in order to observe its "effect".

-Likert Scale--a technique for measuring the strength of respondents' attitudes toward an idea, person, item, etc.  Participants are asked whether they strongly agree, agree, are neutral, disagree or strongly disagree with the statement. Likert scales are subjective.

-Longitudinal research--data are collected for more than one time period, the group may be observed sequentially over specific time intervals. This allows for collection of data before and after changes that may have occurred as a result of the causal factor that was involved in the study.  Analysis involves measurement and assessment of change over time within the same cases, that is, individuals or aggregates. Longitudinal research is more costly due to repeated testing and may suffer from the Hawthorne effect.  See:  Cross-Sectional research as it contrasts with longitudinal research.

-Nominal scale--Scientific measurement consisting of a set of unordered categories, any number can be used as a label. Religious affiliation would be an example, any number could be assigned to each religion such as a social security number, license plate number or a serial code number. The number is just a label with no value attached. The numbers represent different characteristics, e.g., qualities, types or categories (numbers that substitute for names or designations, e.g., Protestant).

-Nonparametric statistics--statistical methods used for analysis of ordinal (higher or lower, more or less type designations) and categorical sample data (categories of things, e.g., diagnoses--heart disease, kidney disease--one diagnoses is not higher or lower in value than the other). They do not require assumptions about the shape of the population distribution from which the samples have been drawn and can be referred to as "distribution-free statistics". They differ in quality but not quantity.

-Objectivity-- Exists independent of our perceptions. Free from distortion, viz., subjective (personal or emotional) bias. (Jary, 1991 p.339) Observations are independent of opinions, feelings and personal reflections. (Powers, 2006 p.115.  See: Subjective

-Ordinal scale--which item or data weighs more-- items are meaningfully numbered from lowest to highest, first, second, third, e.g., Likert Scale.

-Phenomenological research-- inductive and descriptive approach to research reports, the described experience is lived by the person, e.g., coping, distress, acceptance, hunger, love. 

-Primary research--experiments, investigations or tests performed for the first time in order to acquire data, not taken from already published sources. Could also include first hand accounts, diaries, interviews, autobiographies especially in the humanities and history. Primary research differs in various disciplines.  See: Primary, Secondary and Tertiary Research and Resources tab on the first screen.

-Primary research article in the sciences--should include the following sections: abstract or summary or introduction, methods or methods and materials (required), results or data (required), discussion, conclusion, literature cited or references (required).

-Subjective-- the perspective of the person, independent point of view or bias (for) or prejudice (against) something. Once viewed negatively, now both subjective and objective views are thought to support and be necessary to each other. (Powers, 2006 pp. 167-8)

-Theory--any set of hypotheses or propositions linked by logical or mathematical arguments that is advanced to explain an area of empirical reality or type of phenomenon, (Jary, 1991)