Our Process

MFJ begins by collecting data from U.S. statewide court datasets. Often, these populate the majority of our Measures. If feasible, we also approach local agencies (prosecutors’ offices, sheriffs’ offices, public defenders’ offices, etc.) to acquire supplemental data in an effort to ensure we capture the entire criminal justice system. We do this in writing, but also, if possible, in person, via pairs of MFJ researchers who travel our states county by county.

MFJ understands that every county is different. Our process is iterative and responsive to stakeholder feedback. We continually seek input, local context, and information at multiple stages of our work to ensure our data are reliable and take into account county and state differences.

MFJ also recognizes that criminal justice data are not collected, recorded, or systematized uniformly across local jurisdictions. Thus, addressing data quality and uniformity has been a critical MFJ goal from the outset. We have developed a process to clean and code our data in-house and then run them through three audits—two internal and one external—to make sure they are correct. This process involves:

  • Scrutinizing the data to identify data elements that may be unreliable given clear inconsistencies in data collection across counties.
  • Exploring possible inconsistencies between legal and technical language used across and within states.
  • Using text-mining techniques to identify the multiple ways in which the same concept may be recorded using different terms across jurisdictions.
  • Addressing any anomalies, which often reveal that some jurisdictions may not be recording certain events and that, as a result, such events can’t be accurately measured.

Per these reviews, MFJ does not use any data elements identified as problematic.

Our methodology is available for download.

Data Portal

All our case-level data are aggregated to the county level so that the Measures can be calculated and centralized in an online repository—a public Data Portal—that is free and available to anyone. The Portal has been designed so that it’s easy to use but also comprehensive. We provide multiple views to satisfy users of all skill and interest levels (and no personal information is published for any parties involved in a case).

On the Portal, the Measures can be visualized as maps, bar graphs, and tables presenting county-level data. These views are available for download. Users can also compare counties within and across states.

Every Measure will be accompanied by:

  • A short and long definition The long definitions provide some background on the practices being discussed, suggest companion Measures, and note how the Measure was calculated and which cases were excluded.
  • Footnotes The footnotes provide information about individual county practices that are germane to the Measure. We acquire this information from counties or via research.
  • Statutory baselines We include relevant statutes (and links) to explain specific differences in state policies that affect criminal justice (e.g., state A uses sentencing guidelines while state B doesn’t, etc.). This kind of information helps to ensure our data are not misconstrued.
  • Context County context includes information on demographics, population, crime, criminal justice resources, poverty, commute times, voting records, drug hospitalizations, and citizenship.
  • Opportunities to provide feedback Users will be encouraged to send MFJ feedback on every page of our Data Portal. We will update our data and Portal as necessary based on user feedback.

Methodology Summary

Source Data

Measures for Justice (MFJ) works with data extracted from administrative case management systems (CMS). These data are originally collected by the source agencies for the purpose of tracking the processing of individual cases and usually involve manual data entry into the CMS. As such, they may be subject to errors at any stage of the collection and recording process. MFJ excludes unreliable values (e.g., a filing date that is in the future—04/30/2025) and data elements (e.g., the initial appearance date is missing in 80% of cases) from all analyses.

Standardizing Data Across Jurisdictions

Statutory laws, agency practices, terminology, and case management systems vary across and within states. MFJ has developed a Standard Operating Procedure (SOP) to match all data to a uniform coding schema that allows for apples-to-apples comparisons. This includes, but is not limited to:

  • CASE is defined as all charges associated with the same individual defendant that were filed in court (or referred for prosecution, in the case of declinations) on the same date. MFJ assumes that when a prosecutor files multiple charges together, even when they stem from separate incidents, they intend to resolve these charges simultaneously. This may differ from how each agency defines case.
  • CASE SERIOUSNESS is defined by the most serious charge, according to the state’s offense severity classification, that was present at each stage of charging: referral, filing, and conviction.
  • CHARGE DESCRIPTIONS are standardized using a crosswalk that ensures that statutory definitions across states match a uniform code.
  • PRETRIAL RELEASE DECISION represents the court’s initial ruling regarding whether to release the defendant pending case disposition, and whether the release should be subject to monetary or nonmonetary conditions.
  • CASE DISPOSITION indicates the type of action that removed the case from the prosecutor’s or the court’s docket, excluding any actions stemming from appeals or violations of probation. Case disposition categories are defined as follows:
    • Prosecution declined: The prosecutor declined to file all the referred charges.
    • No or unknown disposition: The case was still pending at the time of data extraction or, if it had already been closed, no disposition was recorded in the raw data.
    • Dismissed: All charges that were filed in court were dismissed or withdrawn.
    • Deferred or diverted: The defendant entered a pretrial diversion or deferred prosecution program for at least one of the charges.
    • Not guilty at trial: The defendant was found not guilty of all charges in a jury or bench trial.
    • Guilty at trial: The defendant was found guilty of at least one charge in a jury or bench trial.
    • Guilty plea: The defendant pleaded guilty to at least one charge.
    • Guilty – unknown method: The defendant was guilty of at least one charge but the raw data did not indicate by which method (i.e. trial vs. plea).
    • Transferred: The case was transferred to another jurisdiction. This includes extraditions and changes of venue.
    • Other: Includes other dispositions such as bond estreature and bond forfeiture.
  • TIME TO DISPOSITION is calculated in two ways: (1) the number of days between arraignment and case disposition/sentencing, and (2) the number of days between filing and case disposition/sentencing. For declinations, it is calculated as the number of days between case referral and the prosecutor’s decision not to file. For diversions, it is calculated as the number of days between both case filing and arraignment in court and the defendant entering into a pretrial diversion agreement.
  • ATTORNEY TYPE reports the last attorney of record and includes the following categories: self-represented, private attorney, public defender, court-appointed private attorney, and other.
  • TOP SENTENCE identifies the type of punishment imposed by the court that was the most restrictive of personal liberties according to the following hierarchy:
    • Death penalty
    • Life in prison
    • State prison
    • Jail or county detention facility
    • Lifetime supervision
    • Extended supervision/split sentence with confinement portion in prison
    • Extended supervision/split sentence with confinement portion in jail
    • Extended supervision/split sentence with confinement type unknown
    • Other confinement (e.g., mental health institution, home confinement)
    • Probation
    • Fine
    • Restitution
    • Other (e.g., community service)
    • Time served sentence with no additional confinement time, supervision, or fines.

Data Quality Control

MFJ goes to great lengths to ensure that the data published are as accurate as possible and that the data management process does not become a source of error. MFJ’s data quality control process involves five general stages: (1) assessing the quality and completeness of the raw data delivered by the sources; (2) cleaning the data to remove invalid values and unreliable data elements; (3) conducting several rounds of internal audits of the cleaned case-level data; (4) sending the county-level data out to an independent external auditor to assess the data for face validity; and (5) validating the county-level data with state and local stakeholders.

Measure Calculation

All Measures are calculated at the county level because that is where charging, disposition, and sentencing decisions are made. They are estimated using multiple years of data (five years for most Measures, and two years for those that require controlling for prior convictions) to: (1) increase the number of cases included in the analysis and avoid suppressing smaller jurisdictions that may have few criminal cases on an annual basis; (2) protect the privacy of defendants in small jurisdictions; and (3) reduce the potential effect of temporal instability. The operational definitions, case exclusions, calculations, and sources are provided in all publications of the data.

Data Publication and Suppression Rules

MFJ publishes county-level results on a performance measure only when they conform to the following rules:

  • STATE AVERAGES The counties with available data must represent 50 percent or more of the state’s population for the state averages to be published.
  • NUMBER OF CASES At least 30 cases are needed to generate any performance measure. Performance measures for counties with fewer than 30 cases in the denominator or in the pool to calculate the median are suppressed from publication. Once measures have been filtered by groups (e.g., across race categories), the results are suppressed if the cell contains fewer than 5 cases.
  • MISSINGNESS The maximum permissible percentage of cases with missing values for any given measure is 10 percent. Performance measures for counties with more than 10 percent of cases missing values in the numerator or in the pool to calculate the median are suppressed from publication. In addition, performance measures for counties with more than 5 percent and up to 10 percent of cases with missing values display a “high missing rate” warning.
  • MISSINGNESS BIAS MFJ uses statistical simulations to estimate the amount of bias that may result from missing data. The bias depends both on the percentage of missing data and the actual value of the measure being estimated. For example, in a county where the pretrial diversion rate is low (e.g., 3%) and there is a considerable proportion of cases missing data (e.g., 7%), the estimate of the pretrial diversion rate could be inaccurate. Bias is estimated as a function of the sample mean and the percentage of missing data. Whenever the sample mean and the percentage of missing data suggest a level of bias greater than 5 percent, MFJ suppresses the data from publication.

Disparities

MFJ uses a Relative Rate Index (RRI) to assess disparities in case processing outcomes between whites and nonwhites, males and females, and indigent and non-indigent defendants. The RRI compares how two groups fare on the same outcome by dividing the results of one group by those of the other. An RRI equals to 1 indicates that there is no disparity in outcomes between the two groups. Disparities are not calculated when there are fewer than four cases in the denominator of the rate for either group. We also test the statistical and substantive significance of disparities. Disparities that are neither statistically nor substantively significant are suppressed from publication.

  • STATISTICAL SIGNIFICANCE MFJ estimates confidence intervals to test whether the disparity in outcomes for the two groups is beyond what could be expected by random chance. In this sense, statistical significance provides information about the precision and certainty of the measurement. Statistically significant disparities are noted with an asterisk (*).
  • SUBSTANTIVE SIGNIFICANCE Because statistical significance is affected by sample size, MFJ also evaluates whether the size of the disparity merits attention irrespective of statistical significance. Disparities equal to or greater than 1.05 are considered substantively significant and attempts should be made to understand and address them.


Our measures and methodology have been vetted by two councils of experts: Methods and Measurement Council, and Benchmarking Council. If you have further questions about our methodology, please contact MFJ Research.

User Notes

When viewing the measures users will be asked to keep the following in mind:

  • A Starting PointOur Measures are meant to be a starting point for a conversation about the criminal justice system that addresses what’s working well and what needs further attention. The aim is to create transparency.
  • Adult Criminal CasesOur system measures only the performance of counties on the processing of adult criminal cases. Therefore, we do not measure how juvenile, family, civil, and other cases may fare. Nonetheless, our Measures can be filtered by the age group of the defendant, including those under 18 (juvenile defendants who were waived to adult court).
  • FiltersOur Measures can be filtered by defendant characteristics (race/ethnicity, indigent status, sex, and age) and by case characteristics (offense type, offense severity, court type, attorney type, and drug type–only for drug-specific Measures). We encourage users to explore the Measures using these filters. Some filters calculate disparities between two groups. However, we don't test the statistical significance of such disparities.
  • Data QualityMeasures for Justice (MFJ) works with data extracted from administrative case management systems. These data were originally collected by the sources for the purpose of tracking the processing of individual cases and not necessarily for the purpose of measurement. Nevertheless, they are suitable for measurement provided they are handled correctly. Often, these data are reliable. Just as often, they can be entered incorrectly or not at all, may be subject to errors at any stage of the recording and collection process, and may not be standardized across counties. MFJ has taken steps to account and adjust for these problems but cannot correct entirely for errors in data entry. For these reasons, and because jurisdictions use a variety of calculation methods, we encourage examining overall patterns instead of exact percentages when comparing to reports produced by local agencies.
  • Case DefinitionCriminal justice agencies use different methods to record cases. Some jurisdictions file all charges against a defendant under the same docket number and sometimes they do so even when the charges stemmed from different incidents. Others file each charge under separate docket numbers even when the charges are for the same incident. To standardize the definition of case across jurisdictions, we count all charges associated with the same defendant that were filed (or referred for prosecution, in the case of declinations) on the same date as a single case. We assume that when a prosecutor files multiple charges together, even when they originated from different incidents, they intend to resolve these charges simultaneously. Since the focus of our Measures is case processing, not case clearance, we believe this approach is currently the best way to standardize case definition across jurisdictions.
  • Case SeriousnessBecause cases often involve multiple charges of differing severities, we define cases based on the most serious charge, according to the state's offense severity classification, that was present at each stage of the case processing, respectively referral, filing, and conviction.
  • CausationMFJ’s research is descriptive and does not, by definition, tell us why things happen. As such, we do not test hypotheses about the reasons for the patterns the data reveal. When our Measures show differences between states, counties, or groups (e.g., in medians, percentages, or rates), we make no claim about the reasons for these differences.
  • DisparitiesMFJ uses a Relative Rate Index (RRI) to assess disparities on case processing outcomes between white defendants and defendants of color, males and females, and indigent and non-indigent defendants. The RRI compares how two groups fare on the same outcome by dividing the results of one group by those of the other. An RRI equal to 1 indicates that there is no disparity in outcomes between the two groups. Disparities are not calculated when there are fewer than four cases in the denominator of the rate for either group. We also test the statistical and substantive significance of disparities. Disparities that are neither statistically nor substantively significant are suppressed from publication.
  • Statistical SignificanceMFJ estimates confidence intervals to test whether the disparity in outcomes for the two groups is beyond what could be expected by random chance. In this sense, statistical significance provides information about the precision and certainty of the measurement. When a disparity is statistically significant, we can be 95% confident that the rates for the two groups are unequal. Statistically significant disparities are noted with an asterisk (*).
  • Substantive SignificanceBecause statistical significance is affected by sample size, MFJ also evaluates whether the size of the disparity merits attention irrespective of statistical significance. When a disparity is substantively significant, this means it is large enough to warrant attention. Disparities equal to or greater than 1.05 are considered substantively significant, and attempts should be made to understand and address them.
  • ContextEach Measure sheds light on a corner of a local criminal justice system, but to evaluate the health of that system in a more comprehensive way, all available Measures should be assessed together and interpreted with county context in mind.
  • CountyWe measure criminal justice performance at the county level because it is usually at this level that charging, disposition, and sentencing decisions are made.
  • MissingnessThe maximum permissible percentage of cases with missing values for any given measure is 10 percent. Performance measures for counties with more than 10 percent of cases missing values in the numerator or in the pool to calculate the median are suppressed from publication. In addition, performance measures for counties with more than 5 percent and up to 10 percent of cases with missing values display a “high missing rate” warning.
  • Missingness BiasMFJ uses statistical simulations to estimate the amount of bias that may result from missing data. The bias depends both on the percentage of missing data and the actual value of the measure being estimated. For example, in a county where the pretrial diversion rate is low (e.g., 3%) and there is a considerable proportion of cases missing data (e.g., 7%), the estimate of the pretrial diversion rate could be inaccurate. Bias is estimated as a function of the sample mean and the percentage of missing data. Whenever the sample mean and the percentage of missing data suggest a level of bias greater than 5 percent, MFJ suppresses the data from publication.
  • More DataMFJ continues to seek out more data—especially law enforcement data—as part of our effort to measure all corners of the criminal justice system.
  • TimelineIf you’ve given us data and don’t see them represented in the Portal yet, it’s because we are still working on them to ensure accuracy. Thank you for your participation and patience.
  • Portal UpdatesWe provide a complete history of portal updates that allows you to track when data changes or new data have been released to the portal or when new versions of the portal are made available.