Preface - 1992 Texas School Survey Of Drug And Alcohol Use

-

INTRODUCTION tlse is sponsored by the Texas commission on The Texas school Survey of Drug and AIcohoI Public Policy Resources Laboratory Alcohol and Drug Abuse (TCADA) and implemented 91ttre project is conducted in two parts' (ppRL) at Texas A&M university. The T)xas school survey about every two years in a sample of districts The frst involves a statewide survey administered inform poricymakers about the extent and nature throughout the state. The statewide survey helps gives disrictS conducting local surveys of the substance use problem in Texa, ,rrrooi.. It also own drug and alcohol survey findings' a standard of comp#son for interpreting their makes drug and alcohol surveys available for The second component of the Texas school Survey ^rhe Texas school Survey is offered annually to administration in individuar school districts. partially s9nl.o.ffi tltt"-1it-Ti^t;1eve1 administrations every school district in the state. TCADA the preparation of the executive by absorbing the cost of instrument deveiopment, tunoing as a basis for interpreting local survey data summary of local results, and providing state of Drug and Alcohol Use was findings. A locat administration of the"Texas School Survey your locar survey results are presented of conducted in your school district during Spring ,rgg2. in this repofr along with findings from the statewide survey'

Organization of the RePort part I, District Results, contains a set of tables The report is divided into tfuee major sections. (if applicable) elementary survey data for your and marginals summarizing the ,""orrJuty and volume. It contains a complete set of district. part II, State Results, is bound u, u

"o-p*ion

figurescomparingyourlocalsurveyresultstothestatesurveydata,aswellasasetoftablesand format as youriocal report' Finally' the Executive marginals presenting state results in the same of the iocal survoy resurts. A full report on the Summary contains icustomized written synopsis rcno'L in the fall of 1992' This state sufvey statewide survey proj"a will be published uy in drug use for Texas as a whole' A report will providi a detailed ,o***y of cunent 5e1ds copywillbeforwardedtoyourdistrictaSsoonasitisprinted. should be considered

prior to presenting the results, there are a number of important points which relate both to the 1992 statewide survey with regard to the'data cited in this report. These ittt"j about section briefly presents some basic information as well as to the district-level data. ihis thedevelopmentofthereportandthelimitationsofthedata.

r t_ I

r tl

ll

r l

I l

I

BACKGROUNDoNLocALSURVEYADMINISTRATION of for students in grades four through twelve' Two versions both order and younger students. Districts the instrument have been developed to accommodate we to one of a fixed set of grade combinations' were also asked to limit survey adminisffation instruments and review acceptable will briefly describe the elementary and secondary survey grade combinations below' The Texas school surttey is offered

The Texas School Survev Instrument Two versions of the Texas School Survey are available. The first, a six-page questionnaire, was designed for secondary students in grades six through twelve. All sixth'graders participating in the statewide survey used the secondary survey instrument. However, in local administrations sixth graders could be given the elementary or secondary survey depending on the preference of the district. The secondary survey instrument explored usage patterns of 11 drugs including tobacco, alcohol, inhalants, marijuana, cocaine, crack, hallucinogens, uppers, downers, steroids, and ecstasy. Other questions pertained to behavioral and demographic correlates of substance use, problems relating to substance use, and sources of information about and help for substancerelated problems. students in grades four through six. While questions similar to the secondary survey, the items keep basically was made to the an effort were adapted to be more easily understood by younger respondents. For example, language was simplified wherever possible, students were asked about fewer drug alternatives, and some complex questions were omitted.

A simplified three-page instrument was created for

Standard Grade Combinations

In order to make state and local comparisons of substance use across all grades, both the state and local aggregato percentages must be based upon the same combinations of grades. Otherwise, state and local data will be comparable by individual grade but overall scores cannot be compared directly. Because the state survey data was run in only a limited number of grade combinations, districts were encouraged to administer their local surveys in the same grade groupings. They were as follows: Elementarv

Secondarv

4 through 6 4 through 5

7 through 12 6 through 12 9 through 12 8, 10, artd 12 6, 8, 10, and 12

If your district did not survey one of the above grade combinations, be aware that you can only compilre your district results to the state data by individual grade. Where districts chose to do other combinations of grades, no comparable overall percentages for the state data are available. This will be an especially critical issue for small districts who only receive aggregate data.

l_

I

LIMITATIONS OF THE DATA

t_

When interpreting the state and local survey findings cited in this report, it is important to be aware of a number of subtle factors which can have a major bearing on the meaning of the data. It is prudent to temper conclusions with an understanding of issues such as sampling error and the limitations of self-reported data. It is also important to be sensitive to the precise population to which findings can be generuTized. Knowing who was not incladed in the survey can be as important as knowing who was included. In the following paragraphs, some of the most significant caveats relating to both the statewide and local findings will be discussed.

t_

t_

r t_

r r r t_

r I

r r l. 1_

r t_

Survey Population

In generalizingthe results of the state and local surveys, it is impor0ant to be aware that the data was collected in public schools. Neither private school students nor drop-outs were sampled though both represent significant components of the youthful population of Texas. These limitations should be kept in mind when considering the implications of the data. The findings do, however, represent reasonable estimates of drug use among the public school population of Texas, and offer a suitable basis of comparison for public school districts. Self-Report Data The drug and alcohol use data is entirely based on self-reported information. While a number of studies have estabiished the usefulness of self-reported information for estimating the incidence and prevalence of drug use, the validity of these data ultimately depends on the truthfulness, recall, and comprehension of respondents. This study was carefully designed to minimize the impact of these potential sources of error. For example, students who reported impossibly high levels of substance use or use of non-existent drugs were eliminated from the analysis. Further, differences in recall or comprehension are assumed to be constant across both the state and local survey samples. Any reporting bias in the data should therefore be approximately equal in both surveys and comparisons between district and state data should be valid. Nonetheless, the limitations of self-reported information should be considered in drawing conclusions about the data.

Sampling Error The 1992 statewide survey data presented in this report is based on a sample of Texas secondary and elementary students. Because a sample rather than the entire population of students was surveyed, a certain amount of error will be introduced when generalizing findings from the sample to the population. To gauge the size of this error, confidence intervals for all estimates must be ascertained. A complete discussion of sampling error for the state survey sample, along with confidence intervals for each substance included in the survey, will be available in the state survey report to be produced this fall. In general, however, small differences of a few percentage points will usually not be statistically meaningful. Attention should be focused on large differences of five percentage points or more.

_I ANALYTIC GUIDELINES In processing the drug and alcohol survey data, there were a number of analytic to be addressed' These included questions such as how to identify

issues that had students who were exaggerating their drug or alcohol use, how best to calculate average substance use figures, how to handle surveys from students in grades that were not technicatt-y ueing surveyed and how to classify data for students who failed to report their grade level. This seJtion will review some decision rules estabrished to deal with theie mattersl

Handling of Exaggerators As was discussed above, the survey data presented herein is based entirely upon respondents, description of their own behavior. It is inevitable that some students will under- or over-report their use of drugs or alcohol. To the extent possible we attempt to identify

and eliminate data from those respondents. Two checks were incorporated into the data analysis program to identify exaggerators' First, data from students claiming to have used ozz. a non-existent drug, was considered suspect and dropped from the analysis. Second, students claiming impossibly high levels of drug or alcohol use were also dropprd frothe analyses.l

Weighted Aggregate Data

In many districts, the percentage of all enrolled students that actually take the survey varied substantially by grade. Though patterns are different for each districf rypically a larger proportion of younger

stl dents complete the survey compared to their older peers. If a simple average is taken when calculating overall levels oi'u.", older students may not be represented in proportion to their numbers. sitr." these are the very students most liicely to use drugs or alcohol' their underrepresentation in the aggregate scores would result in artifrcially lower overall levels of substance use.

To achieve more representative percentages when calculating overall or aggregate levels of substance use, a "weighted average" procedure was used. prior to calculating the averages, scores for each grade surveyed were weighted by the total enrollment for that grade. By weighting alleviate d.

averages by actual enrollment, over- or under-sampling

in aggregate scores is partially

Surveys From "Out-of-Range" Grades occasionally, students from a grade that was not technically being surveyed were included in the district sample. This was primarily a problem where the survey was administered in a non'unbelievably high substance use for secondary surveys was defined based on the tbllowing criteria: (l) students reported in item z0 that they had five or more drinks of two or more beverages every day; (2) students reported in item 16 that they used 3 or more alcoholic beverages every day; or (3) students reported in item 13 that they used4 ormore drugs (other than cigarettes, alcohol, oriteroids) ormore times in thepast30

ll

days.

r_

r_

r-

t-

t. t"

r r r r r r

r r r r I

I I I

r I

contiguous grade combination (e.g., 8, 10, and 12). In these instances, the final sample often contained a few students who were in "out of range" grade levels. Generally, data for these students was not thrown out. Rather, it was re-coded and included with the next contiguous grade 1eve1. A conservative approach was used whereby younger students could be grouped with Ih"it old"r peers, but older students could not be re-coded to a lower grade. This prevented artificially high levels of substance use in lower grades. There were two cases in which the "out-of-range" data would be dropped from the analyses rather than being re-coded: (1) if there was no contiguous higher grade with which to combine the data; and (Z) ifthe number of respondents in "out-of-range" students was greater than 10 percent of the grade with which they were being grouped. This latter measure was designed to prevent a large number of "out of range" grades from obscuring the true grade-level values.

Surveys

In Which No Grade Level Was Reported

When students failed to roport their grade level it was not possible to determine unequivocally with what grade these students' data should be analyzed. Where grade level was missing, students' data was retained in the sample and an estimate of grade was made based on his or her age. Students that were at age level for beginning a grade were included with that grade. Grade assisnments were made as follows: Aee 9 10 11

t2 L3 T4 15

16

Il+ If both grade and age were

Grade Level

4th grade 5th grade 6th grade 7th grade 8th grade 9th grade 10th grade 11th grade l2th grade

missing, the data was dropped from the analyses.

CONCLUSION We have present€d an overview of the Texas School Survey project, discussed limitations of the survey daia, and reviewed the key analytic issues. In the next section of this report, attention will be focused on the survey findings. The section will be introduced by a brief discussion of the major issues relating to the presentation of the data. Survey results are then presented in the form of figures, tables, and marginals.

OVERVTEW

OI THE DATA

are summarized in this section of The Texas school Survey findings for your school district figures, tables, and marginals' the report. survey findings *" pr"r"tt"d in three formats: fonlat' allowing for Figures portray both state and local survey results in a "side-by-side" figures, ttrglef all of these could direct comparisons to be made. There are a total of twenty which figures were not be produced for every school district. The rules for determining Presentation, below' generaied for each district are described in Standards of Data

Tables 1-22 illustrates general Three sets of tables are available. The first set containing second set, Tables A1-A3, refers information about substance use patterns in the district. The 1-14, contains prevalence tables specifically to students' uso of alcohol. The third set, Tables drugs or alcohol in the past 30 showing the percentage of students who have used specific tables were available for every days, past school yeai, ana during their lifetimes. All are provided in a companion participating district. Tables showing the same data for the state contain the raw volume. Additional data is presented in the marginals' Marginals ali questions' Bear in mind that if percentages of students who gave each responso option for percentages will represent you,. Oisf,ict under-sampled a-particular g.uO"(t) your overall ,,weighted avorages." As a reiult, the raw percentages presented in your marginals may reflect weighted counts rather than actual raw counts.

for your district' When We have produced all the graphics which could be generated should be considered' In reviewing the survey findings, there are a number ol points which the data and discuss some of the the following p*ugruphs we will provide an overview of findings' Finally' the standards of preseitation used in producing the local survey conventions used in the presentation of results will be reviewed. STANDARDS OF DATA PRESENTATION AIcohoI

of Drug and A wide variety of school districts took part in the T,exas school survey for organizing and IJse. In order to accommodate the diversity of participants, guidelines for handling data from presenting the data had to be established. th.." included standards and figures should be produced small and mid-sized districts and for determining which tables of data for each table and for each. These guidelines win be reviewed bel-ow. The sources figure will be documented as well'

Data Aggregation analyses to

grade-level In some districts; too few students were surveyed to allow for reliable grades was grouped together' Grade be performed. Where this occurred, data from several of valid between-group groupings were established so as to (1) maximize the number one of three possible data comparisons possible and (2) plotect students' confidentiality.

lt_ t_

l_

presentation formats was selected for each district, depending on the number of students per grade. The guidelines used are illustrated in Table I-1.

In districts where the number of students per grade varied widely, a format was selected based on the size of the mnjori4r of grades surveyed. However, if one or more grades fell below a total of 30 students, the next lower level of aggregation was automatically selected.2

l_ l_ Table

t_ t_

HIGH SCHOOL

I-1.

Guidelines for Data Aggregation

DISTRICT SIZE

GRAPHICS F'ORMAT

Small: Fewer than

A single

30 students per

grade

percentage is shown representing all students.

t_

Medium: 30 to 90

l

students per grade

Percentages afe aggfegated in grade groupings (e.g., combining grades 6-8 and grades 9-12).

Large: More than 90

Separate percentages are shown for each grade level surveyed.

l_

students per

grade

t_

l_

ELEMENTARY

Small: Fewer than 90 students per

grade

Large: More than 90

l_

students per

grade

A

separate percentage is shown for all students.

Separate percentages are shown for each grade level surveyed.

T-

t_ t-_

I 1_

2The

only exception to this rule was in the case of medium districts in which some of the grades being grouped together contained fewer than 30 students. As long as the total number of students in the group was at least 90, then the data was presented in the standard format for medium districts (i.e., aggregated in grade groupings).

-

Table I-2. Availability of Figures by District Size and Elementary/Secondary Status

Elementarv

Larqe Small Figure I Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20

./ '/ *J/ *t/ ,/ INS *J/ INS ,/ INS *J/ INS ,/ ,/ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ NQ /r/ ,/ ,/ *J/ ,/ */

NQ NQ

*Jt'

Jl *J/ NQ NQ

Secondarv

Larse Medium Small ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ / ,/ ,/ ./ ,/ ,/ ,/ { ,/ ,/ ,/

,/ ,/ ,/ ,/ ,/ Jl ,/ ,/ ,/ /,/ '/ ,/

,/ ./ INS INS INS INS

,/ ,/ ,/ ,/ ,/

J/

,/

,/ ,/ ,/ ,/

,/ ./ ,/ ,/

,l/

*

,/ ,/

,/ INS

KEY:

,/ *J/

INS NQ

Graphic Graphic survey. Graphic Graphic

is available. is available in a modified form that is not directly comparable to the secondary is not available because of insufficient sample size. is not available because the question on which it is based was not asked.

l,-

f-

f f f f f

r f r r lr r r r r r r

Availability of Graphics There are two factors which determined what graphics were produced for a district: (1) the size of the district, and (2) whether elementary grades were surveyed. This section reviews how these factors affect the tables and figures each district received.

Table I-3. Item Numbers for Questions Serving as Sources of Data for Figures

Elementary Survey Source Item

Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20

KEY: NA

OS

Secondary Survey Source Item

12

t7 t9 I7 t9

11

17

t2

t9

10

16

NA NA NA NA NA NA

I6

11 T2 11

r6 L7

24

3Ia 3lb 32b

aa JJ

27 28

1l/17

a-

23

aJI

11,22123

T9/37 35 OS

NA NA

Figure is not available. Data derived from 1990 NIDA survey

JI

g t. lI a

I I I I I I

Table I-4. Item Numbers for Questions Serving as Sources of Data for Tables

I

Elementary Survey Source Item

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14 Table 15 Table 16 Table 17 Table 18 Table 19 Table 20 Table 2I Table A1 Table A2 Table ,A.3 Prevalence Tables B1-14

KEY: NA

11 11

NA

tUt2

Secondary Survey Source Item

t6 t6 t7 t7lr9

T2

19

l4

20

13

I4

NA NA NA

2l 23

ru20,21

29f 29e

NA

19t30

3lL3

31

NA NA NA

3l2r 32 a,c,e,g 32 b,d,f,h ^^ JJ

L7

35 36

18

NA 22t23 15

NA NA

34 a)t 22 25 26

19

r,3,5,6,7,rU12

Table is not available.

10

1,3,5,7,1'1,r8,19,35

I ll-

r r r r r r r r

District Size. The number of students surveyed in large or medium-sized districts was large enough to produce every available graphic. However, according to the data aggregation rules outlined above, graphics illustrating data by grade could not be produced for districts surveying fewer than 30 students per grade. As a consequence, Figures 3 through 6, all of which show grade-level breakdowns, are omitted from both the elementary and secondary versions of reports for small districts. With these exceptions, all other figures and tables are produced for every district regardless of size. Elementary vs. Secondary Graphics. So that comparisons can be easily made between older and younger students, elementary and secondary graphics have corresponding table and figure numbers. However, they are not always directly comparable. Some graphics are not available for elementary students because they are based on questions not included on the elementary survey. In other instances, elementary graphics are available, but only in a modified form. For example, because only secondary students were asked about their substance use in the "past 30 days," a different time frame (e.g., "past school year") was substituted for the elementary versions of those graphics. Also, while secondary tables and figures provide information about eleven substances, elementary graphics are limited to only four -- tobacco, alcohol, inhalants, and mariiuana. Tables I-2 and I-3 illustrate which tables and figures are available for small, medium, and large districts. They also show differences in graphics for elementary and secondary grades.

Data Sources

It is not always clear from looking at a figure or tabie exactly where the data it represents came from. Table I-4 illustrate the survey items that served as data sources for each graphic

l-

presented.

I

Throughout the report, a number of conventions were adopted to maintain consistency in reporting. Across all tables and figures a common terminology was employed and a standard set of symbols was used. In the following paragraphs, we will describe these terms and symbols. We will begin with a set of general definitions followed by a discussion of specific conventions used in tables and marginals. Finally, we will provide a brief explanation of a few selected graphics depicting relatively complex relationships between variables.

l-

r r

CONVENTIONS USED IN DATA PRESENTATION

Definitions There are several terrns used throughout the report which, though seemingly straightforward, may have some bearhg on how you understand and interpret the data. These are outlined below.

F Alcohol. When the category "Alcohol"

appears in a figure or table, it represents the combined use of beer, wine, wine coolers, and liquor. Each of these alcoholic substances may be referenced individually, as well.

Cocaine or Crack. This reference is found only in the prevalence tables (Tables B1-B14). It refers to students that report using either cocaine or crack, or both cocaine and crack. Since there is some overlap between students who have used cocaine and those that have used crack, the percentages of respondents in the "Cocaine or Crack" category should not be a simple sum of percentages in the individual "Cocaine" and "Crack" categories. It should, however, be at least as large as the larger of the two individual categories.

Illicit Drugs. These

are defined as controlled substances. They include marijuana, powdered cocaine, crack, uppefs, downers, hallucinogens, and ecstasy.

Mariiuana Only. This reference is found only in the prevalence tables (Tables B1-B14). It refers to students that report using marijuana b:ut no other illicit substances. The reference to "Marijuana," in oontrast, includes all students using that substance regardless of whether they used other substances or not.

Prevalence. The percentage of students that report having used a substance or substances. Tobacco. The category "tobacco" includes the combined use of cigarettes and smokeless tobacco. This represents a change from the coding strategy used in the 1988 and 1989 Texas School Surveys which only included cigarettes in the tobacco category. Conyentional Symbols and Format

In addition to figures, there are three types of tables in which data may be presented. These include general substance use tables (Tables l-20), alcohol tables (Tables A1-A3) and prevalence tables (Tables B1-B12). Data is also available in its raw form in the marginals. Here we will present several conventions of data presentation. They apply chiefly to tables and marginals.

The Meaning of an Asterisk (*). When an asterisk appea-rs in a table in place of a it typically means that (a) the actual percentage obtained was less than 0.5 percent or (b) fewer than 10 students responded to that item. Asterisks may also appear if data is not available for a district. This would be the case, for instance, if a district did not survey all the grades in one of the standard acceptable grade combinations (see "standard Grade Combinations" in the previous section). percentage,

Never Used/Ever Used Notations. In Table 1 of the general substance tables and in all alcohol and prevalence tables, a standard approach is used for presenting incidence and prevalence information. An "Ever lJsed" category shows the total porcentage of students who say they have used a substance at all, regardless of when or how many times. A "Never 12

t l_ l_ l_ l_ t_

r r t" r l_

Il_ t_

l_ t_ r_

t_

l_

fIr_

IJsed" category shows the total percentage of students who have never used a substance. The percentages shown in the "Ever [Jsed" and "Never lJsed" categories always sum to 100 percent.

Additional columns then provide further information about those students who say they have "Ever lJsed" the substance. This includes information such as when or how often use occurred. These additional columns always sum to ttre total percentage of students in the "Ever lJsed" category. Conversely, the sum of students in the additional columns and those who say they have "Never fJsed" a substance is also 100 percent. For example, looking at the prevalence tables (Tables 81-14), the percentage of students who used any given substance in the past month, past school year, and prior to the past school year total the number in the "Ever Used" column. That number added to the total percentage of students in the "Never Used" column equals 100 percent.

Marginals. As was noted above, the marginals illustrate the raw percentage of students who responded to each survey item. Frequency, porcent, cumulative frequency, and cumulative percent are shown for each individual response option. The frequency shows the absolute number of students who gave the indicated response. The percent is the frequency converted into a porcentage of all students who answered the question. Again, bear in mind that if your district under-sampled a particular grade(s) your overall percentages will represent "weighted averages." As a result, the raw percentages presented in your marginals may reflect weighted counts rather than actual raw counts. The cumulative scores for frequency or percent show how those numbers increase in an additive fashion across all response options. Cumulative scores are primarily useful in survey questions with continuous response categories (e.g., 1 day, 2-3 days,4 or more days). The cumulative figure makes it possible to quickly determine the number or percentage of students who fall above or below a given cutoff point. The number of students who did not answor the question is indicated in the marginals by a small dot on the frst line of data for each survey item. Surveys that were excluded from analyses as exaggerators are not included in the marginals.

Review of Selected Graphics The majority of figures and tables presented in this report are self-explanatory. Most graphics depict state and local survey findings in a straightforward format. However, there are a few figures and tables which show slightly more complex relationships between variables. The following paragraphs will briefly identify and explain these graphics.

8.

Average Number of Experiences with Inhalants (Grades 7 and t2). The numbers shown in Figure 8 illustrate the average number of times all students reported using specific inhalant substances. The category "Total Experience" indicates the average number of times students have used any inhalant substance, regardless of the type.

Figure

13

Figure 75: Prevalence of 30-Day Use Among Students Saying Variaus Substances Are Very or Not Very Dangerous To Use. This figure illustrates actual levels of current substance use among students who believe particular substances are very dangerous compared to those who feel they are not very dangerous. The percentages shown represent the number of students in each of those categories who report having used the substance at least once during the past 30 days.

Figure 17: Prevalence of 30-Day Substance [Jse Among Secondary Studen* by Parental Aaitude. This figure shows actual levels of students' beer and marijuana use by perceived parental attitude. The "Approve" category is comprised of students who said their parents "strongly" or "mildly" approved of adolescents using those substances while the "Disapprove" category is comprised of students who said their parents "strongly" or "mildly" disapproved. "Neither" and "Don't Know" remain as independent categories. The percentages shown represent the number of students in each of these groups who report having used beer or marijuana at least once during the past 30 days. Figure 20: Prevalence of Lifetime Substance Use Among Seniorc - Natinnal vs. Local. This figure compares substance use among the seniors in your district with actual levels of substance use reported among seniors nationally. The prevalence rates for seniors nationally were derived from the 1990 National Institute on Drug Abuse (NIDA) survey. Table 12: 30-Day Prevalence of Use of Selected Substances by (a) Absences Due to lllness or Other Reasons and (b) Conduct Problems. This table illustrates the relationship between substance use and absenteeism or conduct problems among students. The percentages shown represent the number of students who report having used alcohol, inhalants, or marijuana in the past 30 days. Table 12a shows variation in substance use among students who have not missed school versus those who have missed school from one to four or more days. Table 12b shows variation in use among students who had not experienced conduct problems at school versus those who had experienced conduct problems from one to four or more days. The 1992 state survey data indicated that students who report higher absentism or more conduct problems are also more likely to report having used drugs or alcohol during the past 30 days.

I4