Interpreting the Data

In 1997, 24,505 individuals completed the NHSDA questionnaire. Table 1.1 presents the number of people interviewed for the 1997 NHSDA, within each age and demographic group (unweighted N). In all tabular presentations in Chapters 2 to 13 in which race/ethnicity is used as a variable, the "other" respondents not identified as white, black, or Hispanic were eliminated because the category combines disparate groups with differing patterns of drug use, making it difficult to interpret the data. Education and employment status for those aged 12 to 17 also were excluded from analyses because most 12 to 17 year olds were in school at the time of the interview; therefore, educational attainment and employment status were not applicable measures for this age group. In the tables, adult education and current employment are noted to be not applicable (N/A) for youths (i.e., those aged 12 to 17), and the totals for those two variables refer to the 16,661 respondents aged 18 or older.

Table 1.2 presents the estimated number of people in the surveyed population represented by the 24,505 respondents to the 1997 NHSDA for age groups and demographic groups. All the numbers in Table 1.2 are in thousands and should be read by adding three zeros. Thus, Table 1.2 shows that the 3,871 males aged 12 to 17 who were interviewed in 1997 (see Table 1.1) represent approximately 11,501,000 of their counterparts in the U.S. civilian, noninstitutionalized population. More detailed estimates of the number of drug users in the surveyed population are presented in the 1997 Population Estimates report (OAS, 1998b). The percentage distributions of the surveyed population by age group and demographic group are presented in Table 1.3. This table shows, for instance, that 75% of the weighted sample was white, 11% was black, 10% was Hispanic, and 4% was from other racial/ethnic groups.

Many of the tables in subsequent chapters present data for the demographic groups shown in Tables 1.1, 1.2, and 1.3. The rate of use of various drugs in the lifetime, the past year, and the past month is generally shown for the four age groups by gender, race/ethnicity, population density, and region (and for those aged 18 or older, by educational attainment and current employment status). The prevalence rates for the age groups can be compared in terms of these basic demographic variables. When other demographic variables are used or when data are presented for specific subpopulations, such as current users, the appropriate unweighted N is presented in parentheses. Generally, estimates in the text are rounded to whole numbers; the only exceptions are for estimates that are below 5%.

Most tables of this report do not present estimates of the total numbers of persons using drugs or having other tabulated characteristics. Such population estimates can be readily computed from the tables of this report, however, by multiplying the rate, expressed as a percentage, by the corresponding population base reported in Table 1.2, and dividing by 100. For example, Table 1.2 shows that the estimated number of persons in the surveyed population in 1997 who were aged 26 to 34 and who resided in the Northeast equaled approximately 6,244,000. Table 10.5 shows that the percentage of persons in this subgroup who reported drug use with needles in their lifetime equaled 1.0%. It follows that the estimated number of surveyed population persons in the Northeast aged 26 to 34 who ever used drugs with needles equals .010 x 6,244,000 = 62,440.

All comparisons, as well as the individual rates themselves, are subject to sampling error that is readily quantified. Sampling error for an individual rate results from asking questions of a sample rather than of everyone in the surveyed population. Sampling theory provides the basis for calculating confidence intervals (CIs) around the estimates and tests of significance in comparing two estimates. The size of the intervals and the tests of significance depend on (a) sample size; (b) the interaction between the sampling procedure used and the distribution of a particular variable in the population (i.e., the appropriate design effect); and (c) the degree of confidence required in the interval estimate, or level of protection against incorrect inferences ("Type 1 error"), required in the test of significance.

Standard errors (SEs) of the estimates and associated confidence intervals are not presented in this report, but tables that contain the standard errors are available from OAS. The recommended approach for computing confidence intervals and statistical comparisons would be to obtain and use these standard error tables. Generalized standard errors for various percentages and sample sizes, however, are provided in Appendix C of this Main Findings report. These generalized standard errors and the formulas in Appendix C provide a means of computing approximate confidence limits for various observed estimates at given effective sample sizes at the typical confidence level of 95% (also see Turner et al., 1992).

The tables in Chapter 2 on trends in drug use include an indication of whether a change between 1996 and 1997 was statistically significant. Differences in the proportion reporting use of a drug in 1996 and the proportion reporting use of a drug in 1997 were tested using a differences-in-proportions test. Statistical significance is reported for the .05, .01, and .001 levels in Chapter 2, and for the .05 level in Chapters 3 to 13. A significant change is operationally defined as a change so large in magnitude that it was likely to occur fewer than 5 times in 100 by chance alone (.05 level). Analogous definitions apply to significance at the .01 and .001 levels. Note that the calculations included adjustments that accounted for the design effects of each of the 1996 and 1997 surveys to determine whether differences between proportions were statistically significant.

Nonsampling error, which includes nonresponse, misreporting, and miscoding, cannot be measured as satisfactorily as sampling error. A series of studies on the validity and reliability of general population survey data are reported elsewhere (Turner et al., 1992). The quality control techniques (described in Appendix B) that were used in the NHSDA questionnaire design, field procedures, and data processing are those commonly used to minimize nonsampling error.

 

Demographic Characteristic	Age Group in Years				Total
	12-17	18-25	26-34	35+
Total	7,844	6,239	4,387	6,035	24,505
Gender
Male	3,871	2,756	1,766	2,443	10,836
Female	3,973	3,483	2,621	3,592	13,669
Race/Ethnicity
White, non-Hispanic	3,970	2,966	1,939	3,568	12,443
Black, non-Hispanic	1,346	1,062	1,184	1,047	4,639
Hispanic	2,086	1,863	1,134	1,176	6,259
Other, non-Hispanic	442	348	130	244	1,164
Population Density
Large metro	4,072	3,408	2,301	3,183	12,964
Small metro	2,289	1,813	1,308	1,727	7,137
Nonmetro	1,483	1,018	778	1,125	4,404
Region
Northeast	843	618	684	760	2,905
North Central	992	731	762	770	3,255
South	2,226	1,733	1,765	1,830	7,554
West	3,783	3,157	1,176	2,675	10,791
Adult Education1
Less than high school	N/A	1,580	918	1,474	3,972
High school graduate	N/A	2,254	1,510	1,821	5,585
Some college	N/A	1,842	1,093	1,344	4,279
College graduate	N/A	563	866	1,396	2,825
Current Employment2
Full-time	N/A	2,922	2,856	3,061	8,839
Part-time	N/A	1,257	450	553	2,260
Unemployed	N/A	602	291	248	1,141
Other3	N/A	1,458	790	2,173	4,421

¹ Data on adult education are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
² Data on current employment are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
³ Retired, disabled, homemaker, student, or "other."
Source: Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse, 1997.

Table 1.2 Estimated Number of People (in Thousands) in the U.S. Civilian, Noninstitutionalized Population, by Age Group and Demographic Characteristics: 1997

 

Demographic Characteristic	Age Group in Years				Total
	12-17	18-25	26-34	35+
Total	22,547	27,691	35,246	130,722	216,206
Gender
Male	11,501	13,981	17,490	61,045	104,017
Female	11,047	13,710	17,755	69,677	112,189
Race/Ethnicity
White, non-Hispanic	15,166	18,661	24,449	102,893	161,169
Black, non-Hispanic	3,227	3,874	4,392	12,912	24,406
Hispanic	3,007	3,807	4,643	10,120	21,578
Other, non-Hispanic	1,147	1,350	1,761	4,797	9,054
Population Density
Large metro	9,196	11,261	16,793	53,345	90,595
Small metro	8,059	10,116	11,776	46,497	76,447
Nonmetro	5,292	6,314	6,677	30,881	49,164
Region
Northeast	3,825	4,408	6,244	25,864	40,342
North Central	5,154	6,309	8,593	28,299	48,356
South	8,492	11,147	12,655	48,938	81,231
West	5,076	5,826	7,754	27,621	46,278
Adult Education1
Less than high school	N/A	5,809	4,988	25,785	36,581
High school graduate	N/A	9,974	11,429	43,760	65,162
Some college	N/A	8,969	9,266	27,585	45,820
College graduate	N/A	2,939	9,563	33,592	46,095
Current Employment2
Full-time	N/A	13,285	24,448	65,260	102,993
Part-time	N/A	6,293	3,546	11,328	21,167
Unemployed	N/A	2,316	1,911	4,327	8,554
Other3	N/A	5,797	5,341	49,808	60,945

¹ Data on adult education are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
² Data on current employment are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
³ Retired, disabled, homemaker, student, or "other."
Source: Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse, 1997.

Table 1.3 Estimated Percentage of the U.S. Civilian, Noninstitutionalized Population, by Age Group and Demographic Characteristics: 1997

 

Demographic Characteristic	Age Group in Years								Total
	12-17		18-25		26-34		35+
Total (Row Percents)	10.4		12.8		16.3		60.5		100.0
Gender
Male	51.0		50.5		49.6		46.7		48.1
Female	49.0		49.5		50.4		53.3		51.9
Total	100.0		100.0		100.0		100.0		100.0
Race/Ethnicity
White, non-Hispanic	67.3		67.4		69.4		78.7		74.5
Black, non-Hispanic	14.3		14.0		12.5		9.9		11.3
Hispanic	13.3		13.7		13.2		7.7		10.0
Other, non-Hispanic	5.1		4.9		5.0		3.7		4.2
Total	100.0		100.0		100.0		100.0		100.0
Population Density
Large metro	40.8		40.7		47.6		40.8		41.9
Small metro	35.7		36.5		33.4		35.6		35.4
Nonmetro	23.5		22.8		18.9		23.6		22.7
Total	100.0		100.0		100.0		100.0		100.0
Region
Northeast	17.0		15.9		17.7		19.8		18.7
North Central	22.9		22.8		24.4		21.6		22.4
South	37.7		40.3		35.9		37.4		37.6
West	22.5		21.0		22.0		21.1		21.4
Total	100.0		100.0		100.0		100.0		100.0
Adult Education1
Less than high school	N/A		21.0		14.2		19.7		18.9
High school graduate	N/A		36.0		32.4		33.5		33.6
Some college	N/A		32.4		26.3		21.1		23.7
College graduate	N/A		10.6		27.1		25.7		23.8
Total			100.0		100.0		100.0		100.0
Current Employment2
Full-time	N/A		48.0		69.4		49.9		53.2
Part-time	N/A		22.7		10.1		8.7		10.9
Unemployed	N/A		8.4		5.4		3.3		4.4
Other3	N/A		20.9		15.2		38.1		31.5
Total			100.0		100.0		100.0		100.0

N/A: Not applicable.

¹ Data on adult education are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
² Data on current employment are not applicable for youths aged 12 to 17. Total refers to adults aged 18 or older (unweighted n=16,661).
³ Retired, disabled, homemaker, student, or "other."
Source: Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse, 1997.

SAMHSA, an agency in the Department of Health and Human Services, is the Federal Government's lead agency for improving the quality and availability of substance abuse prevention, addiction treatment, and mental health services in the United States.    Site Map | Contact Us | Accessibility | Privacy Policy | Freedom of Information Act  Disclaimer | Department of Health and Human Services | SAMHSA | White House | USA.gov * PDF formatted files require that Adobe Acrobat Reader® program is installed on your computer. Click here to download this FREE software now from Adobe. What's New Highlights Topics Data Drugs Pubs Short Reports Treatment Help Mail OAS