Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 Sarah Franklin October 30 th, 2013 Disclosure Control for Tables of Frequency Counts using Administrative Justice Data.

Similar presentations


Presentation on theme: "1 Sarah Franklin October 30 th, 2013 Disclosure Control for Tables of Frequency Counts using Administrative Justice Data."— Presentation transcript:

1 1 Sarah Franklin October 30 th, 2013 Disclosure Control for Tables of Frequency Counts using Administrative Justice Data

2 Overview of presentation  In 2013, the Canadian Centre for Justice Statistics (CCJS) placed two administrative crime surveys in the Research Data Centres (RDCs)  Methodologists and subject matter experts developed a scoring approach for tables of frequency counts to identify ‘safe’ tables  Each variable in a dataset is assigned a sensitivity score. A table’s overall score is the sum of the variable scores. If the score is below a given threshold, the table is safe. 2

3 Uniform Crime Reporting Incident- based Survey (UCR) and the Homicide Survey  Administrative datasets  Mandatory reporting by all police services  Criminal incidents substantiated by the police  UCR is a sample of crime data not all crime comes to the attention of the police over 2 million incidents of crime annually  Homicide Survey data more sensitive than UCR All homicides; 543 homicides in 2012  Information on incident, victim(s), accused(s) 3

4 UCR, Homicide variables available to researchers  most serious violation for the incident of crime (e.g., homicide, robbery, mischief)  geography (region, province, CMA)  location (e.g., residential home, convenience store)  weapon causing injury (e.g., handgun, knife)  relationship between victim and accused  age and sex of victim and/or accused  clearance status (accused charged vs cleared otherwise) 4

5 Publicly available STC police- reported crime data UCR and homicide data available to all Canadians:  CANSIM tables (very aggregate)  Tables and graphs appearing in Juristat articles  Custom tabulations upon request 5

6 Homicides by CMA20112012 victimsrate per 100,000 victimsrate per 100,000 Edmonton504.2332.7 Toronto861.5801.4 St. John’s42.100 Montréal541.4471.2 Ottawa111.270.7 Kingston0000 Saguenay10.742.7 Trois-Rivières10.721.3 Sherbrooke10.51 Moncton0000 Québec30.460.8 Brantford21.400 Canada5981.75431.6 6

7 2009 RDC Pilot - Homicide Survey  Homicide Survey was available through RDCs  Results positive, 4 proposals submitted and 3 research reports completed  Researchers commented on the ease of use of data file, documentation and wealth of data/information  Researchers noted that vetting of data tables too long  RDC analysts noted that data disclosure rules difficult to implement and required additional work 7

8  Statistics Act, paragraph 17(1)(b): No person […] shall disclose […] any information obtained under this Act in such a manner that it is possible from the disclosure to relate the particulars obtained from any individual return to any identifiable individual person, business or organization.  Main disclosure issues: Identity disclosure: can identify an individual Attribute disclosure: learn something new  Group attribute disclosure: learn something about a group Residual disclosure: disclosure by combining results 8 Disclosure Issues : What are we concerned about?

9 RDC disclosure control rules for tabular administrative justice data  Scoring approach developed by the Institut de la Statistique du Québec and is used by all STC administrative datasets in the RDCs assign a sensitivity score to each variable table’s score = sum of variables’ scores if table score greater than a threshold value, cannot release table  Go back and use more aggregated variables with lower scores Or perform controlled rounding 9

10 10 Reviewed all variables to appear on the RDC files Identified variables to be excluded due to: unique identifiers name of victim/accused, date of birth of victim/ accused, fingerprint of accused, incident file identifier data quality issues aboriginal variable, firearm information (registered, licensed) too sensitive homicide victim was pregnant, blood alcohol level of homicide victim, person accused of homicide has suspected mental or developmental disorder

11 11 Aggregated sensitive codes of variables Incident clearance status (UCR, Homicide Survey) suicide of accused → cleared otherwise Most serious violation aggregations Homicide Survey 1 st degree murder, 2 nd degree murder → murder manslaughter, infanticide → other homicides UCR sexual violations against children → other sexual assaults

12 12 Scored all UCR variables to appear on the RDC files 0 = not sensitive region=national; sex of victim/accused; vehicle type; target vehicle; motor vehicle recovered; fraud type; property stolen; location of incident; attempted vs completed violation; most serious weapon status 1-7 = sensitive but can be used in a table 8 = sensitive, cannot appear in a table police service id, exact date of incident Table threshold: ≤7 pass; ≥ 8 fail

13 13 Sensitive variables on the UCR, Homicide surveys Variables deemed sensitive (score 1-7)  geography (region, province, CMA)  age of victim/accused (aggregated, detailed)  most serious violation (aggregated)  most serious weapon (aggregated, detailed)  clearance status (aggregated, detailed)  level of injury (aggregate, detailed)  relationship of victim and accused (detailed, aggregated)

14 Detailed relationship between victim and accused (score=4) 14 Homicide victim was killed by: Husband/wifeSeparated husband/wife Common-law husband/wifeSeparated common-law h/w Divorced husband/wifeExtra-marital lover Same-sex spouseex same-sex spouse Father/motherStep-father/mother Son/daughterStep-Son/step-daughter Sister/brotherOther family Close friendOther intimate relationship Authority figureNeighbour Criminal relationshipBusiness relationship strangerCasual acquaintance unknownOther

15 Aggregated relationship between victim and accused (score=3) Homicide victim was killed by:  Family – spouse  Family – parent  Family – other  Other intimate relationship  Casual acquaintance  Criminal relationship  Stranger  Other  Unknown, n/a 15

16 Factors considered when scoring a variable  Scores, thresholds consistent across surveys  Maximum number of dimensions for RDC tables 8 dimensions for UCR; 3 for Homicide  Homicide data: single year vs 10 year data  Wanted scores to work for all CCJS tables: UCR scores: passed all CANSIM and Juristat tables Homicide scores: passed all CANSIM tables but not all Juristat tables 16

17 Factors considered when scoring a variable  Principle behind scoring approach: table is safe as long as sensitive characteristics cannot be attributed to a person or a group  Scrutinized tables with scores < 8 for sensitive characteristics revealed through: Identity disclosure Examined cells with counts of 1 or 2 Attribute disclosure Examined full cells, zero cells 17

18 extract of UCR table with score=7 Sexual violation incidents, victim=female age 25-34, accused=male, Canada, 2011 relationshipWeapon causing injury Unknownphysical force Firearmknifeothern/aTotal friend1310015184 Business1100007486 Criminal0400004 Casual790004190291 Stranger052002218273 Step-parent030001013 Step-child0000033 Other intimate13000812 Neighbour120001922 Total2841704188391,306 18

19 Status of UCR, Homicide RDC pilots UCR:  Crime data for 2007-2011 available in RDCs  7 research proposals submitted and accepted  Disclosure control vetting committee for the pilot ensure disclosure control rules applied correctly evaluate/fine tune disclosure control approaches Homicide:  Homicide data for 1961-2011 available in RDCs 19

20 Pros and cons of scoring  Pros easy for RDC researchers and CCJS to apply rules rules are consistently applied no distortion of data  Cons determining scores and thresholds is time-consuming difficult to determine scores if lots of variables or variables have lots of categories for Homicide, the pass/fail scoring approach for RDCs is very restrictive not immune to residual disclosure 20

21 Conclusion The scoring approach for frequency counts works well:  for crime-reported data and effectively mimics subject matter experts’ judgement when vetting  for census administrative data with an extensive history of published tables that set the standard for releasing tables  when there are a manageable number of variables and categories within variables Once developed, the scoring approach is easy to apply 21

22 For more information, please contact / Pour plus de renseignements, veuillez contacter: Sarah Franklin Sarah.Franklin@statcan.gc.ca 22


Download ppt "1 Sarah Franklin October 30 th, 2013 Disclosure Control for Tables of Frequency Counts using Administrative Justice Data."

Similar presentations


Ads by Google