THE REVISED CONFLICT TACTICS SCALES

1
THE REVISED CONFLICT TACTICS SCALES CTS2
2
1. VARIABLES MEASURED   2. HISTORY OF THE
CTS   3. THEORETICAL BASIS   4. MEASUREMENT
STRATEGY   5. RELIABILITY AND VALIDITY   6. USES
OF THE CTS   7. CRITICISMS OF THE CTS  9. MODE
OF ADMINISTRATION   10 SCORING OPTIONS AND
NORMS   11 HOW TO LOCATE RESEARCH ON THE CTS
3
1. VARIABLES MEASURED BY THE CTS  
CTS2 5 MINUTES FOR 3 CORE SCALES 10-15 MINUES
FOR ALL 5 FOR SELF AND PARTNER       
NEGOTIATION        PSYCHOLOGICAL
AGGRESSION        PHYSICAL ASSAULT         
SUPPLEMENTAL SCALES AND QUESTIONS INJURY SEXUAL
COERCION    CTSPC 6-8 MINUTES FOR CORE
SCALES 10-12 MINUTES WITH SUPPLIMENTAL SCALES
       NON-VIOLENT DISCIPLINE       
PSYCHOLOGICAL AGGRESSION        PHYSICAL
ASSAULT          SUPPLEMENTAL SCALES AND
QUESTIONS NEGLECT DISCIPLINE IN PAST WEEK
4
2. HISTORY OF THE CTS   QUALITATIVE EXPLORATION
(1970-71) ACTS, VOCABULARY, MEANING   FORM A
(1971-72) STUDENT RESPONDENT STUDY ILLUSTRATES
REPLICATION FOR SEVERAL ROLES   FORM N 1975
NATIONAL FAMILY VIOLENCE SURVEY R 1985
NATIONAL FAMILY VIOLENCE SURVEY    

• MANY MODIFICATIONS BY OTHERS
• REVISED CTS
• CTS2, 1996
• CTSPC, 1998
• ADULT-RECALL/CHILD-REPORT VERSIONS, 1999
• SHORT FORMS 2002 (IN PROCESS)
• MANY STUDIES
• 300 IN BIBLIOGRAPHY
• 5 PUBLISHED PER MONTH

5
3. THEORETICAL BASISCONFLICT VERSUS CONFLICT
TACTICS DOES NOT MEASURE CONFLICT, EXCEPT BY
INFERENCE
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
CTS24CHILD-REPORT, ADULT-RECALLAND SIBLING
VERSIONS Role-Specific Versus General
Items Mother, Father, Parents Parent who took
care of you the most
Reporting Period or year  
10
CTS1 VERSUS REVISED VERSIONSADVANTAGES
OF THE CTS1

• ESTABLISHED RELIABILITY AND VALDITY
• NATIONAL NORMS,
• EXTENSIVE BODY OF RESEARCH FINDINGS
• AT LEAST 400 PAPERS

11
ADDVATAGES OF THE REVISED CTS   CTS2        ADD
ITEMS TO INCREASE RELIABILITY VALIDITY       
REVISE WORDING TO INCREASE CLARITY AND
SPECIFICITY        ENTIRELY NEW NEGOTIATION
SCALE REPLACES REASONING SCALE        ADD
SUPPLIMENTAL SCALES INJURY SEXUAL
COERCION        BETTER DIFFERENTIAT MINOR
SEVERE SUBSCALES FOR MALTREATMENT SCALES)
FOR NEGOTIATION SCALE C COGNITIVE AND
EMOTIONAL SUBSCALES        SIMPLIFIED FORMAT AND
VOCABULARY (7TH GRADE)        RANDOM ORDER OF
ITEMS TO REDUCE RESPONSE SET   CTSPC       
MAKE ITEMS MORE AGE-APPROPRIATE        REPLACE
REASONING BY NON-VIOLENT DISCIPLINE       
SUPPLIMENTA SCALES AND QUESTIONS       
NEGLECT        WEEKLY DISCIPLINE  
12
4. MEASUREMENT STRATEGY         BEHAVIOR ACTS
VERSUS INJURIES MALTREATMENT VERSUS
ABUSE DOES NOT MEASURE ATTITUDES OR
MOTIVES INTENDED FOR USE WITH OTHER
INSTRUMENTS          MINOR AND SEVERE
LEVELS          MODULAR AND SYMETRICAL         
FULLY CROSSED EXPERIMENT ANALOGY (3 BY 2 BY
N)       
13
CONCURRENT AND RETROSPECTIVE DATA CONCURRENT
PRECEDING YEAR, MONTH, ETC RETROSPECTIVE ADULT
CHILDREN ACCOUNTS OF PARENT-TO-CHILD PARENT-TO
-PARENT        REFERENT PERIOD       
FREQUENCY OF OCCURANCE RESPONSE
CATEGORIES        SEVERITY ORDER VERSUS RANDOM
ORDER OF ITEMS INTENDED FOR USE WITH OTHER
INSTRUMENTS
14
CTSPC ITEMS AND SCALESPREVALENCE AND CHRONICITY
15
CTSPC
16
CTSPC
17
CTSPC
18
CTSPC
19
CTSPC
20
CTS2 ITEMS AND SCALESPREVALENCE AND CHRONICITY
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
5. RELIABILITY AND VALIDITY   SEE STUDIES LISTED
IN PAPER CTS18        RELIABILITY INTERNAL
CONSISTENCY TEMPORAL         FACTOR
STRUCTURE 
26
(No Transcript)
27
(No Transcript)
28
RELIABILITY OF THE CTSPCINTERNAL
CONSISTENCY (Alpha)       Physical Assault
.55       Severe Assault (Physical
Abuse) .02       Psychological Aggression
scale .60       Nonviolent Discipline
.70       Neglect .22
WHY THE LOW ALPHA COEFFICIENTS?Items measure
rare events Extremely skewed distributions
drastically lowers correlations Do not meet
other assumptions such as equal
variance Multiple forms of physical abuse even
more rare i.e. correlations between items
inherently low
29
INTERNAL CONSISTENCY AND TEMPORAL CONSISTENCY
RELIABILITY          ZERO RELIABILITY ZERO
VALIDITY APPLIES ONLY TO TEMPORAL CONSISTENCY
(TEST-RETEST) RELIABILITY        A SCALE WITH
ZERO INTERNAL CONSISTENCY CAN HAVE HIGH
VALIDITY        TEMPORAL CONSISTENCY FOR CTS1
PARENT-CHILD VERSION .49 (McGuire
Earls,1993) .70 and .79 (Johnston, 1988) .80
(Amato, 1991).   low
30
  VALIDITY HIGH DISCLOSURE
RATE CONTENT VALIDITY CONCURRENT
CORRELATION OF PARTNER REPORTS CORRELATION
WITH OTHER INSTRUMENTS CONSTRUCT
31
CTS RATE
ISCOMPARISON OF CTS WITH GREATERNATIONAL
CRIME VICTIMIZATION SURVEY 18 TIMES
NATIONAL VIOLENCE
AGAINST WOMEN SURVEY 15 TIMESCLIENTS OF
COUPLE AND FAMILY THERAPY 4 TIMES
NATIONAL CHILD ABUSE NEGLECT DATA SYSTEM 16
TIMESNATIONAL INCIDENCE STUDY OF CHILD ABUSE
8 TIMES
32

• WHY DOES THE CTS SECURE
• MORE DISCLOSURE? 
• CONTEXT OF LEGITIMATION (WE ALL DO IT APPROACH) 
•  
• EXCULPATORY INTRODUCTION
• HOW MANY TIMES DID YOU DO IT, NOT WHETHER DONE
•  ASKS WHAT BOTH RESPONDENT AND PARTNER DID 
• FAMILY PROBLEMS CONTEXT RATHER THAN SAFETY,
• INJURY CRIME, OR FEAR 
• FOCUS IS ON BEHAVIOR RATHER THAN OUTCOMES
• SUCH AS INJURY (INJURY IS RELATIVELY RARE)
• AVOIDS DISPARAGING TERMS SUCH AS ABUSE OR
  VIOLENCE 
• ALSO MEASURES NEGOTATION AND PSYCHOLOGICAL
  AGGRESSION, WHCH CAN JUSTIFY HITTING (TRIED
  EVERYTHING)
• DOES NOT DEPEND ON BEHAVIOR BEING KNOWN OR
  REPORTED BY OTHERS

33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
6. USES OF THE CTS   RESEARCH PREVALENCE
RATES THEORETICAL RESEARCH PROGRAM OUTCOME
EVALUATION RESEARCH SOCIAL CHANGE
RESEARCH   CLINICAL SCREENING IDENTIFY
CASES PROFILE OF SCORES ON THE FIVE
SCALES   WITH THE PERSONAL RELATIONSHIPS
PROFILE HUGE NUMBER OF POSSIBLE RESEARCH
ISSUES CASES IDENTIFICATION AND SCREENING FOR
TREATMENT MATCHING   OTHER USES CODING
RECORDS CODING OBSERVATIONS PEER   YOUR USES
???
39
7. CRITICISMS AND LIMITATIONS OF THE
CTS   PURPORTEDLY FEMINIST CRITICISMS       
RESTRICTED TO CONFLICT RELATED VIOLENCE       
IGNORES CONTEXT        IGNORES NON-CONFLICT
ASSAULTS        DOES NOT TAKE INTO ACCOUNT
SMALLER AVERAGE SIZE OF WOMEN        DOES NOT
MEASURE INJURY        EQUATES ACTS THAT DIFFER
IN SERIOUSNESS        IGNORES WHO INITIATES THE
ASSAULTS   OTHER CRITICISMS OF THE CTS       
FREQUENCY RESPONSE CATEGORIES ARE
UNREALISTIC        ONE YEAR REFERENT PERIOD IS
UNREALISTIC        LIMITED SET OF VIOLENT
ACTS        CTS2 LONG IF ALL 5 SCALES ARE
ASKED FOR RESPONDENT AND PARTNER   AS A
PARENT-CHILD MEASURE        STILL NOT WELL
SUITED TO INFANTS ) ONE YEAR REFERENT PERIOD
TOO LONG FOR FREQUENT ACTS
40
9. MODE OF ADMINISTRATION   INTERVIEW
(FACE-TO-FACE PHONE) SELF ADMINISTERED
QUESTIONNAIRE   TIME FOR ADMINISTRATION       
10-15 MINUTES FOR EITHER CTS2 OR CTSPC       
SHORTENING TO 3 CORE SCALES CTS2 7-10 MIN
CTSPC 6-8 MIN   RESPONDENTS        CTS2 -
EITHER OR BOTH PARTNERS, CHILDREN       
CTSPC - PARENTS, CHILDREN        AGE RANGE 10
AND UP AS AN INTERVIEW        NEW PICTURE
VERSION OF CHLDREN UNDER 10        7TH GRADE
READING LEVEL FOR QUESTIONAIRE USE        USE
WITH DIFFERENT ETHNIC AND NATIONAL GROUPS  
41
HIERARCHICAL VERSUS INTERSPERSED
ITEMS CONCURRENT VERSUS RETROSPECTIVE
DATA       CONCURRENT PRECEDING YEAR, MONTH,
ETC       RETROSPECTIVE ADULT CHILDREN
ACCOUTS OF PARENT-TO-CHILD PARENT-TO-PARENT
42
10. SCORING THE CTS   EACH CTS SCALE CAN HAVE
FOUR VERSIONS        FREQUENCY IN PAST YEAR (OR
OTHER TIME PERIOD)        PREVALENCE IN PAST
YEAR        PREVLANCE EVER        CHRONICITY IN
PAST YEAR   SUBSCALES FOR MINOR AND
SEVERE   CTS2 NUMBER OF VARIABLES THAT CAN BE
CREATED        5 TOTAL SCALES, EACH WITH 2
SUBSCALES 15 SCALES        EACH SCALE CAN HAVE
FOUR VERSIONS 15 4 45   CTSPC NUMBER OF
VARIABLES THAT CAN BE CREATED        3 TOTAL
SCALES, PLUS 3 PHYSICAL ASSAULT SUBSCALES
2 SUPPLIMENTAL SCALES FOR NEGLECT AND PAST WEEK
DISCIPLINE 8 SCALES        EACH SCALE CAN
HAVE FOUR VERSIONS 8 4 24   SEVERITY LEVEL
TYPES        NONE, MINOR ONLY, SEVERE  
43
HOW TO TAKE SEVERITY INTO ACCOUNT WAYS OF
MEASURING SEVERITY       MINOR AND SEVERE
BEHAVIOR ITEMS AND SCALES       FREQUENCY OR
CHRONICITY SCORES       SEVERITY X FREQUENCY
WEIGHT SCORE PROBLEMS SKEWNESS OVERLAP 
44
ONE SOLUTION TO SKEWNESS IS PREVALENCE MINOR, SEV
ER, TOTAL PROBLEM MINOR IS NOT PURE
MINOR BETTER SOLUTION IS SEVERITY LEVEL 0
NONE 1 MINOR ONLY 2 SEVERE CORRELATIONS
ARE ABOUT 25 HIGHER THAN WITH PREVALENCE
SCORES CAN ALSO BE USED AS NOMINAL CATEGORIES
IN ANOVA AND MULTINOMIAL LOGISTIC REGRESSION
45

WEIGHTING       FREQUENCY WEIGHT IS PART OF
ABOVE       SEVERITY WEIGHTS NOT USUALLY
ADVISABLE EXCEPT FOR CLINICAL POPULATION      
LENGTH OF REFERENT PERIOD WEIGHTS NOT WORTH
DOINGTO COMPUTE FREQUENCY (CHRONICITY)
WEIGHTS USE MIDPOINTS 3-54, 6-108,
11-2015, 20 25
46
OTHER SCORING ISSUES
SPSS SYNTAX FILE ON WEBSITE         SEND
ANSWER FORMS TO WPS         COUPLE SCORES ARE
AMBIGOUS     BOTH PARTNERS OR BOTH
PARENTS          USING THE NORMATIVE
TABLES (HANDBOOK page 43)
47
(No Transcript)
48
COUPLE TYPES
APPLIES WHEN THERE IS ONE OR MORE INSTANCES OF
THE BEHAVIOR 1 MALE PARTNER ONLY 2 FEMALE
PARTNER ONLY 3 BOTH PARTNERS
49
Fig. 6. Couple Symmetry Types By Sex of
Respondent (All Assaults)
Both Aggress
Both Aggress
Male Aggressor
Male Aggressor
Female Aggressor
Female Aggressor
MALE RESPONDENTS N177
FEMALE RESPONDENTS N376
Chi-square 1.803, p .406 N for Male Respondents
177 (Male Aggressor19, Female Aggressor28,
Both Aggress130) N for Female Respondents 376
(Male Aggressor35, Female Aggressor77, Both
Aggress264)
ID12 E1a
50
NORMS AND CLINICAL INTERPRETATION  NORMS IN THE
CTS HANDBOOK ARE FOR CTS1CTS2 AND CTSPC
TABLES SHOWING MEANS AND STANDARD DEVIATIONS ARE
CRUDE NORMS BUT CTS2 MEANS ARE BASED ON
STUDENTS 2-3 TIMES HIGHER THAN
MARRIED COUPLES CLINICAL INTERPRETATION       
PHYSICAL ASSAULT, SEXUAL COERCION, AND NJURY
SCALES ANY INSTANCE        INDIVIDUAL ITEMS
SHOULD BE EXAMINED IN ADDITION TO THE SCALE
SCORECTSPC ONE INSTANCE OF HITTING IS NOT
LEGALLY PROHBITED ANY INSTANCE OF SEVERE
ASSAULT ITEM
51
SPSS SYNTAX TO CREATE SEVERE ASSAULT SCALE,AND
PREVALENCE, AND CHRONICITYMEASURES OFSEVERE
ASSAULT WHERESA variable name for the
Severe Assault ScaleSAP variable name for
Severe Assault Prevalence ScaleSAC variable
name for Severe Assault Chronicity ScaleITEM1,
ITEM 2, etc variable names for the severe
assault items in the scale 1. RECODE ITEM1 to
ITEM x etc (70). This makes the
response not in last year, but previously, zero
for purposes of computing scores for
the previous year.2. COMPUTE SA ITEM1 ITEM2
ITEM3 etc This sums the items in the severe
assault scale3. VARIABLE LABELS SA SEVERE
ASSAULT SCALE. This labels the new variable 
52
5. COMPUTE SAP SA.Creates a duplicate of the
SA scale in order to transform it into a
prevalence measure6. VARIABLE LABELS SAP 'SEVERE
ASSAULT - PREVALENCE'.Labels the new variable7.
RECODE SAP (1 THROUGH HIGH 1).Gives all cases
with one or more instances of a severe assault a
score of 18. COMPUTE SAC SA.Creates a
duplicate of the SA scale in order to transform
it into aChronicityy measure 9. VARIABLE LABELS
SAC 'SEVERE ASSAULT - PREVALENCE'.Labels the new
variable10. RECODE SA (0 SYSMIS).Classifies
all cases with no violence as missing to create a
variable that applies only to those who engaged
in one or more assaults, and therefore permits
computing the mean number of assaults in the
referent period.
53
11. HOW TO LOCATE RESEARCH ON THE CTS         
TABULAR SUMMARIES OF METHODOLOGICAL ASPECTS OF
THE CTS (1972-1997) PAPER CTS18 ON
WEBSITE          CTS BIBLIOGRAPHY (1972-1997)
PAPER CTS11R AND 11RA ON WEBSITE         
SOCIAL SCIENCE CITATION INDEX TO SEARCH FOR
CITATIONS OF CTS1 STRAUS, 1979 STRAUS,
GELLES, STEINMETZ, 1980 OF CTS2 STRAUS, HAMBY,
BONEY-McCOY, SUGARMAN, 1996 STRAUS, HAMBY,
FINKELHOR, MOORE, RUNYAN 1998
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)