Research Methods and Statistical Analysis Overview

Complete observer

Role in which a researcher makes observations without interacting with those being observed. Participants may not even be aware they are being observed.

Observer as participant

Role in which the researcher does a one-visit interview with participants.

Participant as observer

Role in which the researcher partakes in the daily lives of participants while making their identity and purpose known.

Complete participant

Role in which the researcher partakes in the daily lives of participants but keeps their identity and purpose hidden.

Inductive reasoning

Method that begins with observation and then moves to theories and hypotheses.

Grounded theory

Development of theory based on gathered data/observations.

Convenience sampling

Sampling that is based on what is convenient for the researcher. Also known as 'accidental.'

Snowball sampling

Sampling that is based on referrals from participants.

Content analysis

Scientific, objective, systematic, quantitative, and generalizable description of communications content.

Secondary analysis

Reanalysis of data that were originally gathered for other purposes. Sources include ICPR, Government and private agencies, published and non-published data.

Meta analysis

A systematic review that involves a statistical analysis of the studies' results.

Crime mapping

Use of mapping software to analyze crime incidents. Crime hotspots.

Manifest content

Obvious, clearly measurable items. Recording units. Ex: View military recruiting commercials and record the number of commercials that showed the American flag.

Latent content

Underlying meaning of content. Context units. Ex: View military recruiting commercials but analyze how they appeal to the viewer's sense of patriotism.

(ICPSR) Inter-university Consortium for Political and Social Research

A collection of over 750 academic institutions and research organizations and a data archive of over 250,000 files pertaining to social and behavioral sciences.

(NACJD) National Archive of Criminal Justice Data

A collection of over 3,100 curated studies or statistical data series. Based on crime and justice.

Unit of analysis in a meta-analysis

The individual study.

Policy analysis

Process of researching and identifying potential solutions to a current or anticipated problem.

Program evaluation

Research style that studies the extent to which a program was implemented as planned and produced the desired outcomes.

Process evaluation

A type of evaluation research that involves studying program implementation. Inputs, activities, and outputs. Was the program implemented as intended?

Outcome evaluation

A type of evaluation research that studies the types of outcomes produced by the program's inputs, activities, and outputs. Did the program work?

Evidence based practice (EBP)

The use of rigorous research to inform decisions about which programs are effective and which should be discarded.

Univariate

A type of statistical analysis that examines one variable at a time.

Bivariate

A type of statistical analysis that examines two variables at a time.

Multivariate

A type of statistical analysis that examines three or more variables at a time.

Descriptive statistics

Statistics used to summarize the distribution and relationship among variables.

Inferential statistics

Statistics used to estimate the degree of confidence in generalizations made from a sample to the entire population from which the sample was drawn.

Mode

Most frequently occurring value in a dataset or distribution. Utilized when the most probable value needs to be determined. Ex: 2, 3, 4, 2, 5, 7 (the mode is 2).

Mean

Average of all scores in a distribution or dataset. Utilized for variables measured at the interval or ratio level. Ex: 4+0+2+1 = 7. 7/4 = 1.75 (the mean is 1.75).

Median

Score or point that divides the distribution or dataset in half. Utilized when extremely high or low scores affect the mean. Ex: 11, 8, 4, 9, 10 (the median is 4).

Positively skewed distribution

Tail of distribution extends to the right. Mean is greater than the median.

Negatively skewed distribution

Tail of distribution extends to the left. Mean is less than median.

Frequency distribution

Utilized to report the exact number of cases for certain values and to compare percentages.

Graphs

Provide a picture that is easier to comprehend in comparison to numerically displayed frequency distributions.

Proportion

Value that totals to 1.

Cross-tabulation

Display of distribution of one variable for each category of another variable. Utilized to control one or more variables while examining associations among others.

Percentage

Value that totals to 100.

PRO(S) of Secondary Analysis

Inexpensive, convenient, quality (valid), clean (no error), de-identified

CON(S) of Secondary Analysis

Different units of analysis, undefined variables

Ethical/Moral Dilemmas (field)

Time commitment, Going native, Illegal behavior

Advantages (EBP

Informed decision making, consistency, accountability, reduces ineffective practices

Aspect of CRIJ due to EBP

Intervention? Hot spot policing?