|
Unit of Analysis
| Individual |
|
Universe
| Residential households, from people aged 12 years and older from all states and territories. The scope excluded non-private dwellings (hotels, motels, boarding houses, etc.) and institutional settings (hospitals, nursing homes, other clinical settings such as drug and alcohol rehabilitation centres, prisons, military establishments and university halls or residence). Homeless persons were also excluded as well as the territories of Jervis Bay, Christmas Island and Cocos Island. |
|
Time Method
| Repeated cross-sectional study |
|
Sampling Procedure
| Multi-stage stratified area random sample;
- The sample was stratified by State or Territory with the smaller states of Tasmania, Australian Capital Territory and Northern Territory allocated sample size boosts for reliable estimates. Queensland's sample size was boosted by 1200 as purchased by Queensland Health;
- Each State or Territory, exempting the ACT, was stratified by region as capital city or rest of state;
- Within capital cities Census Collection Districts (CCD) were selected with probability proportional to the number of private households at the 2006 Census;
- Within Country areas Statistical Local Area (SLA) were selected with probability proportional to the number of private households at the 2006 Census. Within the SLA, Census Collection Districts (CCD) were selected with probability proportional to the number of private households at the 2006 Census;
- Within Census Collection Districts the starting address was randomly selected, with interviewing beginning at the dwelling next door, and following dwellings selected via skip intervals, eligible and ineligible addresses, and procedures for blocks of flats/units;
- Within selected households Individuals were selected via the 'next birthday' method. |
|
Collection Mode
| Self-completion |
|
Type of Research Instrument
| Structured |
|
Characteristics of Data Collection Situation
| The 2010 questionnaire was modelled on the 2007 questionnaire, however, an exclusive drop and collect method was used in 2010 for the first time, with computer-assisted telephone interviews (CATI) and drop and collect being used in previous years along with personal interviews. For the first time very remote locations were not deemed inaccessible in all cases, with field workers flown to these locations. |
|
Actions to Minimize Losses
| Strategies used in the 2010 NDSHS to minimise non-response and non-contact include: (1) Fieldworkers call-backs at different times on different day and provision of pre-paid envelope; (2) Protocols used to ensure selected dwellings fully attempted for contact and/or response; (3) Introductory letter from AIHW Director (translated into Italian, Greek, traditional Chinese, Vietnamese and Arabic); (4) Frequently asked questions (translated into Italian, Greek, traditional Chinese, Vietnamese and Arabic); (5) Brochure provided outlining the purpose and importance of the survey; (6) Use of calling cards when required; (7) A AIHW '1800' number for confidentiality questions; (8) A Roy Morgan Research '1800' number for operational questions. |
|
Weighting
| Imbalances due to sampling procedure, response rates, under-sampled states, and over sampled states and territories are adjusted for with the weighting variables for each stratum. Main weighting takes into account geographical stratification, household size, age and sex based on the latest published ABS estimated resident population data |
|
Cleaning Operations
| The data were checked by the archive for missing variable and value labels, out of range values and wild codes, logical inconsistencies, and confidentiality. Some minor changes have been made to variable names and labels where issues were identified in the data cleaning process. The following variable names have been changed since the release of the 2010 National Drug Strategy Household Survey Report: CCDsperStrata=CDstrata, HouseholdsperCCD=HHperCCD, AgeGroupYouth=AgeYouth, AgeGroup1260p=Age1260p, AgeGroup1460=Age1460, AgeGroup1440=Age1440, AgeGroup1460p=Age1460p, AgeGroup1470p=Age1470p, Indigeneity=Indigen, Household=HH, Sexuality=SexOrien, SEIFAquintile=SEIFAqui, BMI_CODED=BMI, TobaccoPerWeek=Tobperwk, CigarettesPerWeek=Cigperwk, EverAnyIllicit=Eanyilli, EverAnyDrug=, EverAnyIllicit1=Eanyill1, EverAnyIllicit2Eanyill2, RcntAnyIllicit=Ranyilli, RcntAnyIllicit1=Ranyill1, RcntAnyIllicit2Ranyill2, RcentAnyDrug=Ranydrug, MonthMari=MoMari, MonthEcst=MoEcst, MonthAmph=MoAmph, MonthCoca=MoCoca, MonthHall=MoHall, MonthInha=MoInha, MonthHero=MoHero, MonthKeta=MoKeta, MonthPain=MoPain, MonthTran=MoTran, MonthSter=MoSter, MonthDone=MoDone, MonthPhar=MoPhar, Guideline2=Guide2, AverageG1=Guide1, Heart_disease=Heartdis, Hypertension=Hyperten, Mental_Health=MentalH, OtherIllness=OtherIll, MaritalStatus=MariStat, BirthCountry=BirthCou, MainLanguage=mainlang, OtherLanguage=OtheLang, Employment=Employ, OtherEmployment=OEmploy, HighestQualification=HighQual, NumberPpl=NoPpl, NumberDepen=NoDepen, Children0to2=Child0_2, Children3p=Child3p, HouseholdIncome=HHincome, Weeks_Preg=Wk_preg, ASGCremoteness_1=ASGCrem1, ASGCremotenessALL=ASGCrem. |
|
Response Rate
| Where response rate is defined as completed questionnaires (26,648), as a proportion of total number of dwellings where contact was made (52,690), the final overall response rate was 50.6%. Of an initial sample of 81,708 households, 2,565 were identified as out of scope, 26,453 could not be contacted and 26,042 were eligible but not willing or able to take part. |
