A study of serious adolescent offenders as they transition to adulthood and out of crime

Interview Information - Subject Release

This measure appears in the following time-points: Release.

Related Construct

• Interview Information

Description of Measure

The Interview Information section of the release interview is comprised of variables that provide basic facts related to when and where the release interview was conducted and items that reflect the subject's current status at the program. Specifically, we include the version number in which the interview was completed, information regarding the location at which the interview was conducted, an identifying code associated with the placement on which the interview was conducted as well as the institution type, and the completion status of the interview.

The following variables provide basic descriptive information related to the interview:

• Study ID [R0CaseID]. Unlike the study ID variable for all other datasets (calendars and measures), the ID for the release interview dataset is a numeric decimal, rather than a whole number. The whole number portion of the study ID is consistent across all Pathways datasets. The decimal portion indicates the order of the release interview in the series of release interviews done by the participant (xx.1 = initial release interview, xx.2 = second release interview for this participant, .3 = third release interview for this participant).
• A marker indicating whether the case was used in the development of the eight dimensions of organizational functioning [R0DimensionID]. More information about the dimensions can be found in the "Dimensions of Organizational Functioning" codebook section of the release interview.
• Interview completion status [R0IntCompStat]. This indicates whether the release interview was fully or partially completed.
• Version number in which the interview was conducted [R0QVersion]. Over the course of the data collection period, the study working group made changes to the wording and/or interview content which sometimes required a programming change. Such changes required the release of a new "version" of the interview. The implication is that some information that is available in interviews conducted in later versions may not be available in earlier versions of the interview. We refer to this as a "version issue" in the Data Issues portion of the measure codebook sections where relevant.
• Interview location [R0IntLoc] (e.g., at the institution on which the release interview is being conducted, at a different institution, public place, etc.)
• Whether the subject is still at the placement at the time of the release interview [R0StillThere]
• Is the program a community-based [R0CommProg]
• Number of different facilities the subject was in during their confinement (e.g., if the subject was transferred between three different facilities - including the release facility - up until their release, this value will be three) [R0FacCnt]
• Status of their involvement with the program at the time of the interview (e.g., program ongoing, completed the program, kicked out of the program, or transferred to a different program) [R0PStat]

An additional set of variables provides more descriptive information about the placement on which the interview was conducted. This includes the length of their stay as well as a link between the release institution and the corresponding stay on the residential services calendar. We recommend referring to the "Out of Community Placements Calendar Documentation" to obtain a better understanding of the residential calendar as well as to become familiar with the calendar terminology (e.g. "gap month") that will be used in this section.

• Identifying facility code [R0PlcCod]. A four-digit code associated with each unique facility, assigned by staff at the Coordinating Center. It should be noted that attempts were made to ensure that unique codes represent distinct facilities; however, it is possible that, due to the participant's limited knowledge, we have inadvertently coded the same facility with multiple codes if various participants provided distinct names for the same facility.
• Institution type (e.g., drug or alcohol facility, jail, adult or juvenile detention, etc.) [R0FacType]. These codes are consistent with the type assigned to the facility in the residential services calendar.
• Recall period month in which the release interview begins and ends [R0Release_Begin], [R0Release_End]. These were plotted using the participant's self-reported date of arrival and date of release from the facility. In most cases this will be in the form of S#M## (e.g., S1M01 for the first month of the follow06 recall period), with some exceptions:
• Before Baseline; indicates the stay began before the baseline interview date.
• L## (S#M#); used when the stay began or ended in a missed timepoint. The L## indicates the linear month number while the S#M# indicates the recall period month that this would have fallen in, if the interview had been completed.
• See CaseID 1228.0 - the stay began in missed timepoint follow12, so R0Release_Begin = L12 (S2M06)).
• L## (gap month); this was used when the stay began or ended in a gap month between recall periods.
• See CaseID 35.0 - the stay begins in a gap month between the end of the follow12 and the start of the follow18 interview. In this case, the stay cannot be mapped to a recall period month number, so R0Release_Begin = L12 (gap month).
• S0M0; indicates the stay began or ended after the baseline interview date but prior to the start of the follow06 timepoint.
• See CaseID 77.0 - the date of the baseline interview is 7/8/02; the stay began 7/24/02; and the first month of follow06 (S1M01) is 8/1/02.
• L## (after end of F84); indicates the stay extends past the end of the follow84 recall period and therefore does not map to a recall period month number.
• Total length of their stay at the release institution [R0Release_NDays]; computed using the self-reported arrival and release dates from the release interview.
• Variables that provide the residential stay number (located in the Facility Stay Series dataset) to which the release interview maps.
• [R0StayNumber_Main] - This will hold the primary stay number from the residential calendar to which the release interview maps.
• [R0StayNumber2], [R0StayNumber3], [R0StayNumber4], [R0StayNumber5] - These will only have a value if the time span of the release stay overlaps with additional stay(s) reported on the residential calendar.
• A marker indicating if the institution on which the release interview was conducted is a match to any stays reported on the residential services calendar in the follow-up interviews [R0Match]. Four codes are used to describe the extent to which the release matches to the calendar; these are described below. A second variable [R0Match_Notes] provides additional information when further clarification is needed.
• 1 = Exact match. The time span of the stay reported in the release interview exactly matches the time span reported in the matching stay from the residential calendar. Since these cases are an exact match to the residential calendar, variables R0StayNumber2 - R0StayNumber5 will have a missing value code of -200 (NA: Release interview is an exact or partial match).
• 2 = Partial match. The release interview matches to a facility stay, but the time period covers more or fewer months than what was reported on the residential calendar. Variables R0StayNumber2 - R0StayNumber5 will have a missing value code of -200 (NA: Release interview is an exact or partial match).
• 3 = Stay match, but overlaps with additional stays that are not a match. The release interview matches to a facility stay, but the time period covered by the release interview overlaps with additional stays reported on the residential calendar. R0StayNumber_Main will provide the stay number to which the release interview exactly matches (e.g., the facility code is a match). R0StayNumberX will indicate the additional unmatched stays. All cases with this code will have additional information the R0Match_Notes field:
• Time covered by release interview spans multiple stays on the residential services calendar. The self-reported date of arrival at the release institution corresponds to their arrival at a different institution immediately prior to the release stay. We believe the subject was reporting their arrival at the start of their initial confinement, rather than their arrival at the release institution. The begin month, end month and total length of stay for these cases is going to reflect their time in both facilities. The release interview, however, correctly reflects their experiences at the latter institution. An example of this (CaseID 202.1) is illustrated below; while the range of months covers stays in both 1019 and 1188, the release interview content focuses solely on facility 1188.

• Release dates incorrectly span two distinct stays in the same facility. The release interview spans two distinct stays in the same facility as indicated on the residential services calendar. The reported arrival falls in the range of months covered by their first stay in that facility, rather than their arrival date for the second stay. The begin month, end month, and total length of stay for these cases is going to cover their time in both facilities. An example of this (CaseID 3588.0) is illustrated below.

• Release interview correctly reflects one stay that appears as more than one on the residential services calendar. In these cases the time period reported in the release interview is correct, but a coding error on the residential calendar resulted in one stay being counted as two. An example of this (CaseID 118.0) is illustrated below. Here, the residential calendar incorrectly reflects two separate stays.

• 5 = No match. The Notes field [R0Match_Notes] will explain why the release interview was not matched to a calendar stay.
• No overlap or match to any stay. The release institution does not appear on any calendar (residential or living calendar) in the follow-up interview. R0StayNumber_Main through R0StayNumber5 will have a missing value code of -97 (Cannot match: No overlap with or match to any stay).
• Entire stay falls in a missed timepoint. R0StayNumber_Main through R0StayNumber5 will have a missing value code of -100 (Cannot match: Entire stays falls within a missed timepoint).
• Stay only appears on the Living Calendar. The release institution was reported only on the living calendar during the follow-up interview. The "Facility Stay Episodes" dataset was created using only residential services calendar data, so these release interviews cannot be matched to a stay. R0StayNumber_Main through R0StayNumber5 will have a missing value code of -99 (Cannot match: Stay was only reported on the Living Calendar).
• Facility name is not a match. The release stay overlaps with a stay reported on the residential calendar, but the facility names are not a match. An example of this (CaseID 52110.0) is illustrated below. In these cases R0StayNumber_Main will indicate the stay number that the release interview overlaps with; however note that these are not an exact match.

• Exact match but incorrect facility code on the calendar. Similar to the example shown above, the release stay overlaps with a stay reported on the residential services calendar, but the facility names are not a match. In these cases, however, the names are a mismatch only due to a coding error on the residential services calendar. An example of this (CaseID 27.0) is illustrated below.

Data Issues

• Caution must be used if the user intends to merge the release database with any other study datasets (calendar or measures) for two reasons. First, we do not have a release interview for every study participant. The release database includes 1130 release interviews across 686 unique study IDs (a single case can have multiple release interviews). Second, the format of the study ID variable for the release database differs from those in all other study datasets in that the release ID (R0CaseID) is in the form of a decimal rather than whole number.
• The decimal in the study ID indicates the order in which the release interviews were conducted (e.g., 35.1 was the first release interview conducted for this ID; 35.2 was the second release interview conducted for this ID). In some cases the numbering sequence will be out of order if an earlier release interview was determined to be invalid and thus removed from the final database. For example, study ID 50236 has three individual release interviews, numbered 50236.1, 50236.3, and 50236.4 (the second release interview for the case was removed so 50236.2 is missing from the dataset).

References

• Questions were developed by the Pathways Study investigators.

This Web site is maintained by the Center for Research on Health Care (CRHC) Data Center. Contact the webmaster at dcweb@pitt.edu. All rights reserved.