TRAINER'S VERSION

CASE STUDY:

ABUSE ON THE INCREASE IN CASCADIA COUNTY

This training version of the case study, Abuse on the Increase in Cascadia County, is based on a case study developed by the Data Users Group of the Oregon Department of Human Services. The lead developers of this scenario were Jim White, Services to Children and Families and Kati Neville, Senior and Disabled Services Division. We thank them and the other members of the Oregon Department of Human Services Data Users for allowing us to use this case study in the SACWIS Supervisory Seminar: Using Information Management to Support the Goals of Safety, Permanency and Well Being Curriculum for Child Welfare supervisors.

Edited by Twin Prime Editorial, Arlington, MA 02474

ABUSE ON THE INCREASE IN CASCADIA COUNTY

You are the program administrator in Cascadia County. Newly appointed Commissioner Smith has asked you to investigate child abuse in Cascadia County. There have been several recent, well-publicized child fatalities. The Capital Courier just published a story about the Child Protective Service (CPS) program and in that article stated that Cascadia County had a 300% increase in the number of abuse and neglect cases last year.

The Commissioner believes that the CPS program in Cascadia is in a state of crisis. The information he has received regarding the fatalities indicates that these children had previous reports made on them but the social workers had done nothing. Furthermore, he believes that child abuse is on the rise in Cascadia because of substance abuse and that boyfriends are the most common offenders. He believes that there are serious problems with the CPS program in Cascadia County and that the workers there are not substantiating enough of the cases they are investigating. This lack of substantiation is creating a situation where children are being left unprotected. He has asked you and the CPS supervisor to visit the State Office for Services to Children and Families (SCF) WebPages and track down the answers to the following questions:

1. (a) How many Child Protective Service reports (CPS) has Cascadia had in each of the past 5 years? (b) How many of these reports were substantiated? (c) In 1998, what percent of CPS reports in Cascadia County were substantiated? (d) How does that figure compare with the state?
   
2. During the period 1992-1998, how does the change in the percent of reports that are substantiated in Cascadia compare to the state's substantiated report percent change?
3. (a) Statewide and in Cascadia County, how many fatalities related to abuse and neglect occurred in 1996, 1997 and 1998? (b) What was the yearly percentage change?
4. What is the age distribution of child abuse victims age 10 and over, both statewide and in Cascadia County?
5. (a) Who is the most common perpetrator of substantiated abuse? (b) How often is it the boyfriend?
6. (a) What have been the top 3 types of abuse over the past 5 years statewide and in Cascadia County? (b) Has there been any notable change?
7. Finally, has substance abuse played a frequent role in the Cascadia cases?

In considering responses to these questions, it is important to look at the data critically. The first step is to locate the data that relates to the question and examine it to make sure that it does indeed answer the question. The second step is to determine if there is additional information related to the topic of the question that should be in the report to the Commissioner.

Question 1: (a) How many Child Protective Service reports (CPS) has Cascadia had in each of the past 5 years? (b) How many of these reports were substantiated? (c) In 1998 what percent of CPS reports in Cascadia County were substantiated? (d) How does that figure compare with the state?

As you get started, it is important to stop and ask what might be the most basic - and most important - questions in the use of data: 'have we found the correct number'?; 'what event or events are most appropriate to answer the question posed?' For example, we will examine data concerning abuse reports and 'substantiated' abuse reports. Which of these is the most appropriate data to look at and report on? It depends on the question that is being asked and requires a good working knowledge of the data, the intent of the question and the definition of the data. The issue of what data to use to respond to a question will resurface again and again.

You and the CPS supervisor look at statistics released by the State Office for Services to Children and Families (SCF) last week. You find data about the number of abuse reports in two tables, Table 1: Abuse Reports - Statewide and Table 2: Abuse Reports -Cascadia. (You can find these tables in Attachments #1 and #2 located at the end of this case study.)

In Table 2: Abuse Reports - Cascadia, you find the answers to Question 1a and 1b. Looking at the column entitled "# of CPS Reports", you see the number of CPS reports for the past five years and in the column entitled "# of Substantiated Reports" you find the number of substantiated reports. With a pen highlight the answers on Table 2.

The numbers themselves, however, don't tell you much. You would expect that the total number of abuse reports would be greater than the number of substantiated abuse reports and that the state would have more reports than the County. You decide to look at the third part of the question to determine what percent of Cascadia's total reports are substantiated and then compare Cascadia's percent of substantiated reports to the state's percent of substantiated reports.

Recalling your basic math, you remember that a percent tells you how many out of 100. The way to calculate percent is to write a fraction whose denominator (the number on the bottom) is the number after the word of and whose numerator (the one on the top) is the other number and then multiply by 100. Frequently you are dividing the smaller number by the larger number. In this case, the formula is:

Number of Substantiated reports * 100
Total Number of CPS Reports.........

Using data from Tables 1 and 2 and the formula above, calculate the percent of substantiated reports for both Cascadia County and statewide in 1998 and compare the results. Note your answers in Answer Box #1 below:

Question 2: During the period 1992 - 1998, how does Cascadia's substantiated report percent change compare to the state's substantiated report percent change?

Tables 1 and 2 give you another way to compare the number of abuse reports and the number of substantiated abuse reports for the state and Cascadia. You notice that in both tables there are columns titled Yearly % Change of CPS Reports and Yearly % Change of Substantiated Reports. Using the formula below, you can see how to calculate the percent change in total number of CPS reports and in substantiated reports.

Percents may be used to show a change--either an increase or decrease-- in an amount, quantity or value. To calculate PERCENT CHANGE, use the following formula:

(Most Recent Number - Previous Number) * 100 = Percent Change
Previous Number........................................

Looking at Table 1 once again, you can see that it indicates that child abuse reports are up statewide. In 1998, 31,456 reports of suspected child abuse were recorded--a percent change of + 12.31 since 1997.

31,456 - 28,008 * 100 = 12.31%
.......28,008

Since 1992 there has been an increase of + 44.15%.

31,456 - 21,822 * 100 = 44.15%
....... 21,822

Look at the same data for Cascadia so you can compare the Cascadia percent change with the state's. You observe that while statewide abuse reports have increased in the last year, reports in Cascadia decreased from 1158 to 1105 during the same time period (see Table 2). This decrease in Cascadia reports represents a -4.58% change since 1997. Even though Cascadia reports were down last year, over the past 6 years (1993-1998) the number of reports has grown. Since 1993, abuse reports in Cascadia have increased 18.18%. Notice how the time period displaying the percent change over many years is different in the narrative than in the Tables. Statewide the time period discussed is 1992-1998 and in Cascadia it's 1993-1998.

When comparing percentage change across entities, it is important to use the same time period.

Now that you have analyzed the number of CPS reports, you need to calculate the percent change in the number of substantiated reports in the state and in Cascadia during the same period so you can answer the Commissioner's question.

Using information from Tables 1 and 2 and the Percent Change formula, calculate the percent change for substantiated reports for the state and Cascadia County for the period 1992- 1998. Compare the results. Put your answers to Question 2 in Answer Box #2 below:

Using percent change data is one way of comparing child abuse reports and substantiated reports in the state and in Cascadia. Note that Tables 1 and 2 also use a rate per 1000.

Definition of Rate and Importance of Rate

A critical issue in attempting to compare data is knowing that the data are really comparable. In many instances, simply knowing the number of events, in this case the number of abuse reports, is not sufficient. For example, we know that there will be more abuse in the State as a whole than in Cascadia - one of its Counties. In order to make these raw data comparable, statisticians calculate rates. This means that the number of events that occurred is compared to the population for which that event could have occurred. The figure is then standardized to some number (such as 100, 1,000 or 10,000.) In fact, when you calculated the percentage of CPS reports in Cascadia County that were substantiated, you standardized to the number 100. (Remember that percent tells how many out of 100. You could also have standardized to some other number, for example, the number of abused children in the county compared to the number in the state.)

You and the CPS supervisor have now found or calculated the data that answers the Commissioner's first two questions. Look over the data you have been analyzing, is there any other information that you think would be useful to the Commissioner that should be included in the report? If so, note the information in Answer Box #2a below.

Question 3 (a) Statewide and in Cascadia County, how many fatalities related to abuse and neglect occurred in 1996, 1997 and 1998? (b) What was the yearly percentage change?

So far, the abuse profile of Cascadia looks better than the state profile. The statistics you have calculated do not even come close to those reported by the Capital Courier. You have double-checked the calculations you have made as well as those you received from SCF and cannot find how the newspaper calculated last year's growth in the number of abuse and neglect reports as 300%.

You decide to move on to answer the child abuse and neglect fatalities question. Complete Table 3 below by calculating percent change in abuse and neglect fatalities for both the State and Cascadia County for the years 1996,1997 and 1998.

Table 3: Abuse and Neglect Fatalities

Do the results look strange to you? Why? (See the Tip below.)

Small numbers, as seen in Table 3, can cause large fluctuations in statistics that can render them virtually meaningless. When confronted with a situation like this, report actual numbers.

Look over the data on Abuse and Neglect Fatalities you have been analyzing. Is there any other information that would be useful to the Commissioner that should be included in the report? If yes, record the information in Answer Box #3 below.

Question 4: What is the age distribution of child abuse victims age 10 and over, both statewide and in Cascadia County?

To answer this question, you and the CSP supervisor decide you need to calculate the percent of child abuse victims age 10 and over for both Cascadia and the State using data from Table 4. To determine the percent of victims age 10 and over statewide, use the Cumulative Percent column in Table 4, subtract the percent of child victims who are age 10 or over (78.52%) from 100.00% which gives us 21.48%.

Since all numbers with a % sign represent a rate per 100, the % numbers in the same data set can be added, subtracted, multiplied or divided as you would other numbers. For example, 2+ 5 =7 and 2% +5%=7%.

Make the same calculation to find the percent of victims who are over age ten in Cascadia and note your answers to both questions in Answer Box #4 below.

Study Table 4, do the percent of victims age 10 and over statewide and in Cascadia seem significant? Is there data that would be more significant to report to the Commissioner? (Consider what percent of child victims are under the age of 1.) If so, note in Box #4a below.

Question 5. (a) Who is the most common perpetrator of substantiated abuse? (b) How often is it the boyfriend?

While anecdotes and personal experience often help provide richness and depth to supplement data, too often they can be misleading and do not actually represent the true nature of the phenomenon being studied.

Tables 5 and 6 (appearing as Attachment #4 at the end of the case study) display the numbers of alleged substantiated abuse perpetrators over the past five years for the State and Cascadia. Unfortunately, the tables you received from SCF do not report any additional categories, such as significant other, boyfriend, or girlfriend.

You and the CPS supervisor plan to make a note of this data gap in your report to Commissioner Smith. Looking at Tables 5 and 6, what data do you think would be important for the Commissioner to know? Consider who are the most likely perpetrators of child abuse in the State and in Cascadia? Is it important to determine what percent of perpetrators are mothers or fathers or what percent of perpetrators are family members?

Look at the table below. To determine abuse by parents, you can add the entries for mother and father. Note that statewide the percent of alleged perpetrators who are mothers or fathers is 67.5%; for Cascadia the percent is 67.2%. Likewise, if you want a single statistic for the percent of abuse perpetrated family members, you can add the percentages for mother, father, sibling, and other relatives

Calculate the percent of abuse perpetrated by family members in Table 7 below. (Remember that you can add the raw numbers or percents of several categories to come up with new categories. Keep in mind that you can do this only if the categories don't overlap. For example if, as in the data reported below, there is only a single alleged perpetrator reported for each case.)

Table 7: Alleged Perpetrator of Substantiated Abuse - 1998

Look over the data you have been analyzing on alleged perpetrators, is there any other information that would be useful to the Commissioner that should be included in the report? If so, note in Answer Box #5 below.

Question 6 (a) What have been the top 3 types of abuse over the past 5 years statewide and in Cascadia County? (b) Has there been any notable change?

Looking at the data in Tables 8 and 9 (appearing as Attachment #5 at the end of the case study), you and the CPS supervisor review the data to determine the top 3 types of abuse in each of the past 5 years. You notice you are missing data both statewide and for Cascadia in the 1994 and 1995 columns for 'Threat of Harm'. You contact the SCF analyst you had been working with to get the missing pieces and he tells you that the data do not exist; the Division only began reporting 'Threat of Harm' data in 1996.

The major question which needs to be asked in a situation where there are new categories for a variable - in this case, 'threat of harm' added as a 'type of maltreatment' - is its impact on the other categories. Is it a subset? Is it pulling numbers from one or more of the other categories?

In examining the data in Tables 8 and 9, the addition of 'threat of harm' in 1996 seems to have a definite impact on the other categories of maltreatment. It appears to be 'pulling' from the other categories. The totals in the other categories are basically on a downward trend as the 'new' category is increasing.

Decide what answer you would give to the Commissioner's questions on types of abuse and note your responses in Answer Box #6 below.

Question 7: Finally, has substance abuse played a frequent role in the Cascadia cases?

Your interview with the SCF analyst revealed that the Commissioner's last question regarding the role of substance abuse cannot be answered with SCF's computerized data. The data SCF has on substance abuse are from a statewide study that cannot be reported by county.

Decide what answer you would give to the Commissioner's questions on substance abuse and note your responses in Answer Box #7 below.

You and the CPS supervisor have now gathered and analyzed the data needed to address the Commissioner's questions. Quickly go back through the answers you have recorded in each answer box to make sure that the data you looked at indeed answer the questions asked and, where you think appropriate, that you have provided additional data, information and explanations in response to the question. Note any changes in the corresponding answer box.

Attachment 1

Table 1: Abuse Reports - Statewide

Attachment 2

Table 2: Abuse Reports - Cascadia

Attachment 3

Table 4: Age of Victims - Statewide and Cascadia

*Percentages don't add to 100% because of rounding.

Attachment 4

Table 5: Alleged Perpetrator of Substantiated Abuse, State

Table 6: Alleged Perpetrator of Substantiated Abuse, Cascadia County

Attachment 5

Table 8: Type of Maltreatment, Statewide

Table 9: Type of Maltreatment, Cascadia County