top of page
Screenshot 2023-07-24 ABel Screening_edi

Interpreting AASI Data

AASI Sexual Interest Graph


Steps for using the AASI VRT Graph to look for high sexual interest in children

  1. Isolate the 8 bars of the appropriate ethnicity: For Caucasian or African-American clients, use the corresponding scores; for Hispanic or Asian-American clients, use the Caucasian scores; for other clients, use the ethnic group with the highest self-reported interest. Each bar has two numbers with it – the first number is the VRT z-score and the number in parentheses is the average self-reported interest.

  2. Looking only at the 8 bars isolated in step 1, there will be a vertical line drawn perpendicular to those 8 lines. If any child sexual interest bar Female 5 or less, Female 6-13, Male 5 or less, or Male 6-13 touches or crosses that line, strongly suspect high sexual interest in that category.

 

Steps for using the AASI VRT Graph to look for high sexual interest in violence

 

  1.  If the S&M AWF exceeds the Female Adult bar by at least one standard deviation, you should consider it a red flag for the possibility of an interest in sexual violence against females and an area that should be explored more closely.

  2. If the S &M AWM exceeds the Male Adult bar by at least one standard deviation, you should consider it a red flag for the possibility of an interest in sexual violence against males and an area that should be explored more closely.   


NOTE: Based on studies conducted in 2005, Abel Screening concluded that while an analysis of those VRT images may be instructive from a clinical viewpoint, caution should be used in reporting these data.


The Probability of Past Child Sexual Abuse Behavior Score for AASI-3 Only
The Probability of Past Child Sexual Abuse (CSA) Behavior score for males and females is designed to measure how much a denying client’s answers match the pattern of a child sexual abuser. The score is useful for clients who have been accused, but deny sexually abusing a child. The higher the score the more likely the client is to match the denying child sexual abusers.

There are separate models for males and females.  Each model was constructed by comparing the AASI test results of denying child sexual abusers to those of community volunteers.  A common measure of accuracy for a predictive model such as these is the area under the receiver operator characteristics curve (ROC).  The ROC of the probability measure for males is 0.83 and the ROC for females is 0.79.

The scores range from approximately -3 to +5.  They can be interpreted as z-scores based on the sample of community volunteers.  A score of 0 represents the average score of a community volunteer.  A score of 1 represents one standard deviation above the mean for a community volunteer or approximately the 84th percentile.  A score of 2 represents two standard deviations above the mean or the 98th percentile of community volunteers.

The Probability of PAST CSA Behavior was developed using both extra-familial and incestuous sexual abusers of girls and/or boys (17-years-old or younger) and is therefore applicable to abusers of both boys and girls as well as incest child sexual abusers and extra-familial child sexual abusers.

 

How to Use Probability Scores:

  1. Probability scores should only be used from the client's first AASI Assessment.

  2.  Use only for:

​                 a. Clients accused of sexually abusing a child where the client denies ever having sexually abused a child.
                 b. Clients who admit to or are accused of possession/distribution of child pornography but deny sexually abusing a child.
                 c. Clients who admit to or are accused of traveling to meet with a minor or communicating with a minor for sexual purposes                              but deny sexually abusing a child.

   

    3. Scores are divided into two groups.  For Males scores above 1.4 are classified as higher risk and those below 0.21 are classified as l            lower risk.  For Females scores above 1.23 are classified as higher risk and those below 0.15 are classified as lower risk.


   4. The Probability of PAST CSA Behavior score should be used only as a guideline and incorporated into the totality of information about the case. Probability scores can never determine guilt or innocence.  The Probability of PAST CSA Behavior score is attempting to classify a client as having sexually abused a child (under 18 years of age) sometime in the past. The score is not attempting to determine the guilt or innocence of your client for any specific accusation.

Note: Cut points for Low/Medium/High were assigned as follows: zero was the low category due to the overwhelming frequency of zeros; the remaining range was then divided into two based on frequency; the lower of those two ranges was labeled Medium and the higher was labeled High.


Emerick Sexual Victimization Scales Cut Scores
The Emerick Scales were derived from The Sexual Trauma Inventory (STI). The STI is a 300-item true-false questionnaire that is constructed to measure a sexual abuse survivor's 1) knowledge about human sexuality, the paraphiliac, and victimology, 2) process of victimization, 3) trauma potentiators, 4) vulnerability to sensory stimulated intrusive memory, and 5) cognitive, social, and sexual symptomology that are related to sexual abuse. The inventory is an assessment instrument which clarifies issues that are challenging to a client's health; however, it is not designed to validate allegations of sexual abuse.

The Emerick Scales measure the relationship between one's own victimization and one's sex offending behavior. High scorers on the Emerick Scales are offenders whose clinical issues include significant trauma relating to their own victimization.

  1. Trauma Intrusive Thoughts Scale – Measures the degree to which the client involuntarily remembers or thinks about the child molestation he or she suffered.

    2. Trauma Symptomology Scale – There are three critical ways that a sexual abuse victim may misinterpret the process of abuse:                     cognitive distortions about violence, justifications enabling violence, and faulty beliefs about the world in general. Collectively,                          these misinterpretations about one's own victimization are a measure of one's trauma relating to abuse and one's potential to hold              enabling views about sex offending behavior.
    3. Trauma Potentiators Scale – Measures the degree to which developmental and social issues are likely to aggravate trauma                           symptomology. This scale comprises both static and dynamic issues.

Slightly more than 50% of the child molesters were sexually abused during childhood. Of the 52 child sexual abuse survivors, 60% successfully completed primary intervention. Whereas only 36% of the 48 child molesters who were not sexually abused during childhood successfully completed primary intervention and complied with probation terms. Chi Square Analysis shows the relationship between one's developmental history including child sexual victimization and one's discharge status from this intervention is significant (x2 (1, n = 100) = 5.855, p < .01). Among hebephiles, 31% of the 26 men who were not sexually abused successfully completed primary intervention and complied with their probation terms. Comparatively, 64% of the 19 men who were sexually abused, successfully completed primary intervention and complied with their probation terms. Chi Square Analysis shows the relationship between one's developmental history including child sexual victimization and a hebephile's discharge status from this intervention is significant (x2 (1, n = 45) = 8.0917.283, p < .001). Among Pedophiles, this relationship is not observed.


Approximately 25% of the child molesters are 25 years old or younger. Of the 26 youthful offenders, 31% successfully completed primary intervention. Comparatively, 52% of the child molesters who were 26 years old or more, successfully completed primary intervention and complied with their probation terms. Chi Square Analysis shows the relationship between one's age at the time of risk to the community and one's discharge status from this intervention is significant (x2 (1, n = 103) = 3.503, p < .05).

bottom of page