Chapter 9: Secondary Data

Primary data collection

Data collection; original data collection

Secondary data collection

Data collection reanalysis of data collected by someone else. 

Ex: news, NCVS

Why conduct research using secondary data?

1. Critically important information resource for researchers

2. Collected using sound methodological techniques

3. Saves time, money, and other valuable resources

4. Breadth of data found may be unmatched with the primary

What is one major advantage of secondary data?

Saves time, money, and resources.

Which is NOT an advantage of secondary data?

Researcher control over measurement

Disadvantages of using secondary data

1. May not be able to address specific research question

2. May not code variables or have the required response categories

3. Researcher may not understand how original data were gathered

4. May not provide nuance about topics

A researcher wants to measure “fear of police,” but the dataset only includes a general trust question. This illustrates which disadvantage?

Inability to code variables appropriately

Which of the following is a disadvantage of using secondary data?

May not include needed response categories

What are the four types of secondary data?

1. Surveys

2. Official statistics 

3. Official records

4. Other Historical documents (e.g., photos)

What is crime mapping used for?

Identifying spatial distribution of crime and linking it to social indicators like poverty. (e.g., neighborhood, census tract)

What is a primary function of crime mapping?

Visually communicating patterns of crime.

1. Provides visual and statistical analyses of the spatial nature of crime and other events 

2. Connects crime data to other data sources, like census information on poverty or school information

3. Provides map to visually communicate analysis results

What is Big Data?

Very large computer-readable datasets used to reveal patterns and associations.

What is a major ethical concern with Big Data?

Loss of anonymity and increased surveillance potential.

Ethical issues in using big data

1. Must maintain subject confidentiality

   • Example: ICPSR-this is where the IPH data came from that you read for the quantitative coding exercise (it is restricted through the CDC)

2. Must examine data for disclosure risk

3. Must ensure protection of human subjects

4. Concerns with Big Data

What is content analysis?

A systematic method for coding and analyzing text to identify patterns and relationships.

Content Analysis

1. Coding and categorizing text

2. Discovering relationship among constructs identified

3. Statistical analysis of those findings.

A researcher codes newspaper articles for themes about domestic violence. This is:

Content analysis

Why is secondary data limited methodologically?

You have no control over how data were collected or measured.

Because you did not collect it so you are working with what you have.

Methodological Limitations of secondary data

1. What were the original goals in collecting the data?

2. Who collected the data?

3. What data were collected and what were they intended to measure?

4. When was the information collected?

5. What methods were used for data collection?

6. How is the information organized? Is the information identifiable?

7. How consistent are the data with available data from other sources?

8. Is there any missing data? What documentation is available?

TL;DR: You have no control over this data because you did not collect it so you are working with what you have.

Which question should ALWAYS be asked when using secondary data?

What were the original goals of the study?

Ethics associated with secondary data analysis:

1. Some secondary data analysis projects require IRB review.

2. Most of the public use data are exempted from review.

3. If subjects' private information is included in secondary data study, review is needed.

4. Identifiable information is often stripped from the file when the data are archived.

5. Federal regulations specify that existing data may be exempted from IRB review.

6. Research agencies and organizations may provide data files with restrictions on usage.

A researcher downloads a public-use dataset with all identifiers removed. What is most likely true?

The study may be exempt from IRB review

A dataset includes names and medical records of participants. What is required before analysis?

IRB review and approval.

Which of the following statements is TRUE regarding ethics in secondary data analysis?

Some secondary data require IRB review depending on content

What is mixed methods research?

Intentional combination of quantitative and qualitative methods.

What is triangulation?

Using multiple methods to study the same research question

No one method is superior to the others and each varies in its suitability for research

only use if it makes sense

Convergent parallel design

In mixed-methods research, when quantitative and qualitative methods are implemented at the same time; the findings are integrated and interpreted together.

Explanatory sequential design

The quantitative method is implemented first, followed by the qualitative method. (quantitative first, then qualitative.)

Exploratory sequential design

The qualitative method is implemented first followed by the quantitative method.

Embedded design

In mixed-methods research, when the primary method is qualitative or quantitative, but the researcher adds the other component to gain additional insight.

Transformative design

In mixed-methods research, this design uses a theoretical perspective with a social justice focus to improve the well-being of vulnerable populations. (e.g., feminist researcher or participatory action research.

Multiphase design

In mixed-methods research, this design involves a series of quantitative and qualitative designs; each design and the findings inform the next phase.

In which design are quantitative and qualitative data collected at the same time?

Convergent parallel design.

Which design collects quantitative data first and qualitative second?

Explanatory sequential design.

Which design collects qualitative data first and quantitative second?

Exploratory sequential design.

Which design embeds one method inside another primary method?

Embedded design

Which design uses a social justice framework?

Transformative design.

Which design involves multiple phases where each informs the next?

Multiphase design.

What is meta-analysis?

Quantitative analysis of findings from multiple studies.

-Enhances generalizability

-Eligibility criteria for literature: distinguishing features, research respondents, key variables, research methods, cultural and linguistic range, time frame, publication type

What is meta-synthesis?

Qualitative analysis of findings from multiple qualitative studies.

What is the key difference between meta-analysis and meta-synthesis?

Meta-analysis is quantitative; meta-synthesis is qualitative.

What is a strength of mixed methods research?

It allows researchers to combine methods to best answer the research question and improve validity.

What is a limitation of mixed methods research?

It is not inherently superior and may be unnecessary or overly complex for some research questions.

Ethics and mixed methods

1. Researcher's role: considerations

   1. Typically, quant researchers see themselves as experts who design a research project and collect data using objective procedures that are best carried out without participant involvement

   2. Typically, qual researchers see themselves as engaging with the research in some type of collaboration with community/group they are studying, with input from research participants

2. Potential for different reactions of participants to different roles (e.g., why unique methods like pláticas are important)

3. Renegotiation of consent at every stage

Which statement best reflects an ethical consideration in mixed methods research?

Participants may respond differently depending on the researcher’s role