7.3: Data Quality Concerns

Criminal Justice Research in Practice: An Applied Guide for Future Practitioners · LibreTexts

0%

Contents

    Text

    Size
    100%
    Theme
    Typeface
    Line length
    Spacing

    Read aloud

    How

    Speed: 1×
    Light up
    Highlight

    Symbols: Mulberry Symbols, © 2018–2026 Steve Lee, CC BY-SA 4.0 (licence, the set), the site's own copies.

    From LibreTexts, under CC BY-NC-SA 4.0, as the source states it; its equations are written in TeX, which the PDF shows typeset. This page at LibreTexts.

    7.3: Data Quality Concerns

    The quality of research findings depends on the quality of the data used in a study. Even a well-designed study can produce misleading results if important groups are left out, if too few people participate, if crimes go unreported, or if information is recorded inaccurately. As evidence-informed practitioners, it is important to think critically about the quality of the data on which research findings are based.

    Sampling Bias

    Earlier in this chapter, you learned that researchers often use a sample to learn about a larger population. Ideally, the sample should represent the population the researcher hopes to understand. Sampling bias occurs when the people or cases selected for a study do not adequately represent the larger population from which they were drawn.1 For example, imagine a researcher wants to understand community attitudes toward law enforcement but collects responses only from people who attend neighborhood watch meetings. Individuals who attend these meetings may have different experiences, concerns, or opinions than community members who do not attend. As a result, the findings may not accurately represent the views of the entire community.

    Sampling bias can occur in many ways.2 Some groups may be more likely to participate in a study than others. Researchers may unintentionally recruit participants from only certain locations or backgrounds. In some cases, people who choose to participate may differ in important ways from people who choose not to participate. When evaluating research, practitioners should consider whether the sample appears to reflect the population being studied or whether certain groups may have been left out.

    Nonresponse and Missing Information

    Even when researchers select a strong sample, not everyone who is asked to participate will actually participate. Some people may refuse to respond, while others may begin a survey but leave questions unanswered. As a result, researchers may end up with less information than they originally planned to collect.3 For example, a researcher may randomly select 500 correctional officers to complete a survey, but only 150 officers may choose to respond. If the individuals who responded differ from those who did not respond, the findings may not accurately reflect the views of the larger population.

    Missing information can also occur when participants skip questions or provide incomplete answers.4 If large amounts of information are missing, researchers may have difficulty drawing accurate conclusions. As informed consumers of research, it is important to consider not only how participants were selected but also whether enough of them actually participated to provide meaningful information.

    Response Bias

    Surveys and interviews rely on participants providing accurate information. However, researchers must also consider the possibility of response bias, which occurs when participants provide answers that do not accurately reflect their true beliefs, attitudes, experiences, or behaviors.5 For example, individuals may underreport illegal behavior, overreport socially desirable behaviors, or provide answers they believe the researcher wants to hear. As a result, survey and interview responses may not always reflect what actually occurred.

    Hard-to-Reach Populations

    Some criminal justice populations are considered hidden or hard to reach because barriers make it difficult for researchers to identify, contact, recruit, or collect information from them.6 Hard-to-reach populations may be difficult to locate, difficult to contact, reluctant to participate, or difficult to interview because of factors such as privacy concerns, fear of legal consequences, stigma, lack of trust, language barriers, low literacy, or other communication challenges. For example, a researcher interested in studying illegal drug use may have difficulty identifying and recruiting participants because many individuals may not want to admit engaging in criminal behavior. Similarly, researchers studying people experiencing homelessness, gang-involved individuals, victims of domestic violence, or undocumented immigrants may face challenges locating potential participants, gaining their trust, or encouraging them to participate in research. When certain groups are difficult to reach, they may be underrepresented in research findings, which can affect how well the results reflect the larger population.

    The U.S. Census Bureau notes that populations may be hard to reach for several different reasons.7 Some groups are hard to locate, meaning researchers have difficulty identifying or finding members of the population. Others are hard to contact because they may not have stable housing, reliable communication methods, or a fixed location. Some groups are hard to persuade because they are reluctant to participate in research due to mistrust, privacy concerns, fear of consequences, or competing demands on their time. Finally, some populations are hard to interview because of language barriers, low literacy levels, disabilities, or other communication challenges. Understanding these barriers helps researchers recognize why some populations may be underrepresented in research findings.

    Underreporting

    Hard-to-reach populations can also contribute to underreporting. Underreporting occurs when incidents occur but are never reported to authorities or recorded in official data systems. Individuals who are reluctant to interact with researchers or criminal justice agencies may be less likely to report crimes, victimization, or other experiences. As a result, important information about these populations may never be included in research studies or official crime statistics, making it more difficult for researchers and practitioners to fully understand the issues affecting them.

    This is one reason the National Crime Victimization Survey (NCVS) is so important. The NCVS helps researchers understand crimes that may never appear in official law enforcement records. When reading research that relies on official crime statistics, practitioners should remember that official data often reflect reported crime rather than all crime that occurred.

    Data Entry and Recording Errors

    Criminal justice data is only as accurate as the information entered into criminal justice records and databases. Errors can occur when reports are completed incorrectly, information is entered into a records management system inaccurately, incidents are coded improperly, or important details are omitted.

    For example, if an officer records the wrong offense type, enters an incorrect date, or leaves out important information from an incident report, that error may become part of the official record. Researchers, administrators, and policymakers who later rely on those records may unknowingly use inaccurate information when making decisions or conducting research.

    Why Accurate Data Entry Matters

    As future criminal justice professionals, you will not only use data, but you will also help create it. Every time an officer completes an incident report, a dispatcher enters information into a records management system, a correctional officer documents an inmate's behavior, a probation officer records a violation, or an evidence technician enters information into an evidence tracking system, they are creating data that may later be used by supervisors, administrators, policymakers, and researchers.

    The quality of criminal justice research depends on the quality of the information entered into these systems. If reports contain errors, important details are left out, or data are entered inconsistently, those problems become part of the records that researchers may later analyze. As a result, inaccurate or incomplete data can lead to inaccurate or incomplete research findings. Even the most skilled researcher cannot produce high-quality evidence from poor-quality information.

    As you prepare for careers in criminal justice, think about the importance of accuracy every time you complete a report, document an incident, update a case file, or enter information into a database. The information you record may affect decisions made by investigators, prosecutors, judges, correctional administrators, agency leaders, and researchers. Careful and accurate documentation is not simply an administrative task. It is an important part of professional practice and helps ensure that future decisions are based on the best information available.


    1. (Blackstone, 2012).
    2. (Blackstone, 2012).
    3. (Blackstone, 2012).
    4. (Blackstone, 2012).
    5. (Maxfield & Babbie, 2025).
    6. U.S. Census Bureau. (2019, July). Counting the hard to count in a census. U.S. Department of Commerce. https://www.census.gov/content/dam/Census/library/working-papers/2019/demo/Hard-to-Count-Populations-Brief.pdf
    7. (U.S. Census Bureau, 2019).

    This page titled 7.3: Data Quality Concerns was last modified on Tue, 03 Nov 2026 03:24:39 GMT and is shared under a CC BY-NC-SA 4.0 license and was authored, remixed, and/or curated by Denise Borchert, Western Technical College.