Studied by 0 people
Looks like no one added any tags here yet for you.
What is secondary data analysis? Its benefits and disadvantages?
act of compiling or analyzing data that
were originally collected by someone else at another time
Types of Secondary Data
• Official statistics (Census, UCR, NCVS)
• Official records (arrests, sentencing reports, case files)
• Surveys
• Historical documents (newspapers, news media, television broadcasts,
etc.)
BENEFITS: merge different data sets, allows researchers to answer different types of research questions
LIMITATIONS: data quality, data collected for different objective so poor measurement, time period might influence data, ethics & freedom of information act
What is ICPSR and why is it relevant to secondary data analysis?
Inter-University Consortium for Political and Social Research
stores data and information for nearly 7,000 sources and studies, including independent studies and those conducted by
the U.S. Government.
What is comparative research? What types of research questions does
comparative research address?
research that compares data from more than one period and/or more than one nation or geographical unit
Descriptive Comparative Research: research that seeks to understand the structure,
nature, or scope of a nation’s criminal justice systems or rates of crime
Analytic Comparative Research: research that seeks to understand how national
systems work and the factors related to their operations.
Transnational Research: explores how cultures and nations deal with crime that
transcends their borders
EXAMPLES:
The Comparative Crime Data File collected data on homicide rates of 110 nations from
1900 to 1970. Using this data, the researchers were able to test a hypothesis that during
post-war periods, nations experience increased homicide rates (legitimation of violence
model).
Weis, Testa, and Santos (2018) used WHO data from 85 countries, to examine how
drinking patterns influenced homicide rates. Hazardous drinking patterns was
significantly related to homicide, while controlling for total alcohol consumption by
country, and a host of other control variables.
What is a content analysis? What types of research questions does a content
analysis address?
Content analysis is a research method for systematically analyzing and making
inferences from text
Useful for research questions such as
◦ Does media accurately portray crime?
◦ How are drugs marketed via music?
◦ How has media today covered the opioid crisis differently than the use of marijuana during the war on drugs?
blend of qualitative and secondary data analysis. The data is coming from words, or non-numerical sources, and has previously been created or collected by another person, for another purpose. HOWEVER, content analysts normally code their data for quantitative purposes
What is the process of content analysis?
helps to understand about popular culture, and many other issues by studying the messages delivered through mass media and other sources
Establish a theory., rationale, or research question.
Conceptualize the variables that will be used.
Operationalize the variables that will be used.
Create coding scheme and codebooks (human vs. computer coding).
Identify a target population of documents.
Sample the documents if necessary.
Coding training/intercoder reliability
Reporting the results
needs assessment
type of evaluation research that attempts to determine the needs of some population that might be met with a social program
-generally completed using a combination of survey and official data.
-These tools help determine who is eligible for a program, and what types of accommodations they may need
impact/outcome evals
Analysis of the extent to which a treatment or
other service has the intended effect
-This issue normally comes back to causality. Did the program (IV) have an
impact on the target output/outcome (DV).
-Remember, to ensure that a quality impact evaluation is performed, we must
first ensure that the program was implemented as intended
process eval
evaluation research that investigates the process of service
delivery.
-how is the program being implemented? Is the program following the specified rules, has staff been trained, are participants receiving the proper elements, etc.
-examine coverage/how well a program is reaching its intended target population.
-Many times we conduct an impact
evaluation to determine if a program is effective. If the program shows no benefits, we
consider the program ineffective. But that may not actually be the case, instead the lack
of results may have been because the program was not actually implemented correctly.
-normally conducted with qualitative methods, or follow a standardized checklist that ensures proper program delivery
evaluability assessment
type of evaluation research conducted to determine whether it is feasible to
evaluate a program’s effects within the available time and resources.
– is an evaluation of the program even possible?
Why might a program not be evaluable?
– Management only wants their performance to be confirmed, and do not care about
the program having a legitimate effect.
– Staff do not trust outside researchers to check on their performance.
– There is no clear indication of what the program is trying to accomplish.
– The program is not clearly distinct from other services delivered by the agency, and
therefore cannot be evaluated on its own
-easy way to identify evaluability is to create a logic model, indicating exactly what
inputs, outputs, and outcomes are expected from the program. Qualitative interviews are
also useful for indicating whether a program is evaluable
efficiency evaluation
type of evaluation research that compares program costs to program effects
Two types
– Cost-Benefit: specific programs costs and procedures compared to economic value of outcomes.
– Cost-Effectiveness: direct comparison to the program’s outcomes in a non-economical sense (how many jobs were obtained, how many people recidivated, etc.)
-Primary challenge with these evaluations in CJ is quantifying the monetary value
of certain outcomes. How much does incarceration actually cost? What is the
cost of a crime? What is the value of a prevented victimization?
What is the difference between a black box and theory driven evaluation?
A Blackbox evaluation is when an evaluation ignores and does not identify the
mechanisms by which the program produced results
A theory driven evaluation uses the program theory to evaluate the program
What types of research designs (i.e., experimental, quasi-experimental, non-
experimental, qualitative) are best suited to each evaluation type?
impact—experiments
process—qualitative
needs—non-experimental
evaluability—qualitative and non-experimental
efficiency—qualitative and non-experimental
What is a systematic review?
are a summary review about the effectiveness of a program/intervention in which the researcher compares completed studies about
the program, assessing differences across different designs and samples. Often statistical techniques are included, known as meta-analyses
What is data cleaning?
process of checking data for errors after the data have been entered into a computer file been entered into a computer file
Data cleaning is important to:
-Correct errors in data entry
-Identify outliers or unusual cases
- Examine missingness in the data
-Ensure coding has been properly conducted
data is entered in a data set as variables, each variable should have a description and corresponding code
-if using multiple items to measure a concept we create the scale within the data and check it for errors
What is a frequency distribution used for?
numerical display showing the number of cases and using the percentage of cases corresponding to each value or group of values of a variable
-how OFTEN OR FREQUENT a response was selected & useful for identifying percentages of responses
What is variability?
refers to the extent to which cases are spread out through the
distribution or clustered in one location
What are measures of central tendency?
refers to values around which cases tend to center, how much a variable varies across the sample
mode: most frequent value in a distribution
median: the point that divides a distribution in half “middle”
mean: arithmetic average of all scores in a distribution (probs most important in all of inferential statistics
variance: variability of distribution
range: highest data-lowest data
IQR: first and third quartile
SD: square root of average squared deviation of each case from mean
What is positive or negative skew? Can you identify positive or negative skew given an example?
extent to which cases are clustered more at one or other end of distribution instead of around middle
postive-tail to right
negative-tail to left
In a normal distribution, how much of a sample falls between two standard deviations of the mean?
95.4%
Distinguish between univariate, bivariate, and multivariate analyses.
univariate-description of one variable
bivariate- shows relationships between two variables
multivariate-shows relationships between three or more variable
When are crosstabulations appropriate for bivariate analyses?
shows the distribution of one variable for each category of another variable
-nominal and ordinal
nom= cannot be added
ordinal= rank can’t add but one greater than other
use gamma measure of association or chi square
When is correlation appropriate for bivariate analyses?
interval/ratio
use pearson’s r
What is statistical significance? Is it the same thing as the strength of an association?
reflects PROBABILITY that an observed relation is not just due to chance
-DOES NOT MEAN STRENGTH OF RELATIONSHIP OR HOW IMPT RELATIONSHIP IS
What are the advantages/limitations of qualitative data analysis?
-focus on meanings rather than quantification
-Collection of many data of a few cases, rather than few data on many case
-Study in depth and detail without predetermined categories or directions
-The researcher is an instrument rather than a designer.
-Sensitivity to context rather than universal generalizations
-Attention to the impacts of the researcher's values
-A goal of rich descriptions of the world
limited in generalizability, small sample size, time consuming
What is progressive focusing?
process of iteratively analying data, interacting with it, gradually refining the study
What is tacit knowledge?
“unknown knowledge” credible sense of understanding social processes that reflect the researcher’s awareness of participants actions/words and what they fail to state or feel deeply
What is reflexivity?
how YOU influence data and vise versa
What is grounded theory?
revolving process, inductive (based on observations that are summarized into conceptual categories), very repetitive
collects data —> form concepts from data —> group concepts into categories —> identify relationships between categories —>
What is conversational analysis?
focuses on the sequence and details of conversational interaction and on how reality is constructed
What is thematic coding?
A method where researchers identify and label recurring patterns or themes within a dataset by assigning codes to specific sections of text. The purpose is to discover overarching themes by analyzing words and their meaning.
initial coding/line-by-line coding
each sentence is coded individually without a prior assumptions about the meaning or connection to categories
pattern coding
taking initial codes and then grouping them together based on their similarities to create a higher level of analysis (a means of creating deeper meaning in the data
thematic coding
a method where researchers identify and label recurring patterns or themes within a dataset by assigning codes to specific sections of text (purpose is to discover overarching themes by analyzing words and their meaning
How does qualitative research relate to creating theory inductively?
ask about tmrw
