The World Mental Health Composite International Diagnostic Interview
The World Mental Health Survey Initiative
Health and Work Performance Questionaire
NCS: Answers to Frequently Asked Questions

Q: Should I weight the data and which weights should I use?

The data file ncsdxdm2 contains both a Part I (plfwt) and a Part II (p2wtv3) weight. Use weights whenever analyzing the data. The Part I weight should be used for analyses that examine the full sample. The Part II weight should be used whenever the sample is comprised of just those respondents who answered Part II of the survey. The weights are a combination of the various weights described in our papers to adjust for differential household size and differential non-response and post-stratification. We are not releasing the parts of the weight separately, as Rod Little and his students are still doing methodological analyses of these components. These separate component weights will, however, be released in subsequent versions of the data.

Q: Do I need to adjust for the complex sampling?

We also added variables STR (1-42 sampling strata) and SECU (1-2 random halves of each stratum) for purposes of computing design-based standard errors. We are not providing consultation on how to do this, as the methods are complex and we don't have enough staff to teach people about them. However, for users who are familiar with resampling methods, these two variables are all you need. If you are unfamiliar with these methods, you should consult a statistician. You might also want to take a look at the Sage Publications monograph on resampling methods.


It occurs to us that there may be people working on the same topic in different centers. In an effort to help provide a way to coordinate and reduce duplicate efforts, we are hoping you will write in and declare yourself as working on various topics. Then people who are working on similar topics can find each other and get together for purposes of coordination.

Please send us information about your project to:

A request

Please send us a copy of each paper you write using the NCS. We would appreciate it if you could mention in your acknowledgements section that the NCS is funded by the National Institute of Mental Health (Grants R01 MH/DA46376 and R01 MH49098), the National Institute of Drug Abuse (through a supplement to R01 MH/DA46376), and the W. T. Grant Foundation (Grant 90135190).

Please send it to:
Ronald Kessler, PhD, Harvard Medical School, Health Care Policy
180 Longwood Ave, Boston, MA 02115

Q: What are the references for the personality measures in the NCS?

Interpersonal Style
L11-L15 Emotional reliance on others Short form of Hirschfeld (1977)
L16-L17 Assertion of autonomy Short form of Hirschfeld (1977)
L20-L24 Social self-control Short form of Hirschfeld (1977)
Other Personality and Cognitive Dimensions
L1-L4 Internal locus of control Short form of Levenson scale (1973)
L5-L8 External locus of control Short form of Levenson scale (1973)
L9-L10 External locus of control
(powerful others)
Short form of Levenson scale (1973)
L30-L34 Self-esteem Subset of Rosenberg self-esteem scale (1965)
L26-L29 Self-criticism From Blatt et al. DEQ scale (1976) & Rosenberg (1965)
L35-L41 John Henryism Short form of James scale (1983)
L42-L44 Fatalism From Schwartz & Robinson (1991)
L45-L46 Belief in a just world From Lerner (1980)

Blatt SJ, D'Afflitti JP, & Quinlan DM (1976). Depressive Experiences Questionnaire. Yale University Press: New Haven, CT.

Hirschfeld RM (1977).  A measure of interpersonal dependency. Journal of Personality Assessment, 41(6), 610-618. 

James SA, Hartnett SA, & Kalsbeek WD (1983).  John Henryism and blood pressure differences among Black men.  Journal of Behavioral Medicine, 6(3), 259-278. Lerner MJ (1980).  The belief in a just world.  New York: Plenum Press. Levenson H (1973).  Multidimensional locus of control in psychiatric patients.  Journal of Consulting & Clinical Psychology, 41(3), 397-404. Rosenberg M (ED.) (1965).  Society and the Adolescent Self-Image.  Princeton, NJ: Princeton University Press.

Schwartz S & Robinson MM (1991).  Attitudes toward poverty during undergraduate education.  Journal of Social Work Education, 27, 290-296.

Q: What is the reference for the self-description measures in the NCS?

R1-R10 Extroversion From Goldberg (1992)
R11-R20 Neuroticism From Goldberg (1992)
R21-R29 Openness to Experience From Goldberg (1992)

Goldberg LR (1992). The development of markers for the big-five factor structure. Psychological Assessment, 4, 26-42.

Downloading NCS data

The NCS data are archived by the Inter-university Consortium of Political and Social Research (ICPSR) at the University of Michigan. ICPSR is the largest behavioral science data archive in the world. Most major universities in the U.S. and many in other parts of the world are members of ICPSR and have a local ICPSR Official Representative. If you know this individual, he or she can help you with accessing the NCS data. If you do not know the name of your local ICPSR representative or if you do not know if your organization is a member of ICPSR, you can call ICPSR User Support at 313-763-5010. ICPSR staff will provide limited help in downloading the data without charge even if you are not part of a member organization.

Please visit the NCS Data page for more links on this topic.

Q: What type of scale was used to measure socioeconomic status in the NCS?

In general, we use various types of SES: family and personal income, education, ratio of family income to official poverty standards for each family size (from the early 1990s), exogenous variables merged on from Census or PUMS data such as tract level average of median income, percent of tract living below poverty line, and average number of people employed in a geographic area of choice.

Q: Disorders are referred to as "with heirarchy" and "without heirarchy." Could you please inform me as to what this means?

For every disorder we took into account the organic exclusion criteria (i.e. not due to the direct physiological effects of a substance or a general medical condition). However, for some disorders, in addition to the DSM-III exclusion mentioned above, there are other exclusions. For example in MDD criteria C there has never been a manic episode, a mixed episode or a hyponmanic episode. We refer to this as hierarchy. In other words, the condition is not best accounted for by another disorder. When we apply the hierarchical rules we refer to the disorder as "with hierarchy". Please refer to the diagnostic documentation (working paper 7) for a description of hierarchical criteria.

Q: What is the reference for the social support measures used in the NCS?

Aseltine, R.H., Jr., & Kessler, R.C. (1993). Marital disruption and depression in a community sample. Journal of Health and Social Behavior 34(3), 237-251.

Q. Can you describe the terms "narrow, broad, and intermediate" which were used to code 12 month alcohol/drug dependence and abuse?

We followed a specific coding scheme for 12 month alcohol/drug dependence and abuse outlined below. Please note there are broad and narrow definitions for dependence and broad, intermediate, and narrow definitions for abuse.

For 12 Month alcohol or drug dependence:

1. broad dependence requires lifetime alcohol/drug dependence and any dependence symptoms during the past year
2. no intermediate dependence
3. narrow alcohol or drug dependence requires lifetime dependence and at least 3 dependence symptoms during the past year

For 12 Month alcohol or drug abuse:

1. broad abuse is 12 month abuse and no broad dependence during past year
2. intermediate abuse is 12 month abuse and no narrow dependence during past year
3. narrow abuse is 12 month abuse and no narrow dependence during past year and no lifetime dependence