Data intelligence · automation · analyticsA Praxis Global Institute platform
Praxis Data Lab

Praxis Data Lab · Study Builder

Build your NHANES study with harmonization intelligence.

Combine variables from the indexed Continuous NHANES catalog, choose the cycles you need, and see how Praxis connects source history, methods, cycle boundaries, and cross-code lineage into a study-specific harmonization view.

3variables selected
Praxis harmonization guidance available

Praxis found harmonization conditions for this study and has surfaced them below with the supporting source evidence.

3 variable assessments

ALD260
Evidence developinglow confidence

Questionnaire

[During the past 30 days] how often did {you/SP} drink hard liquor such as tequila, gin, vodka, scotch, rum, wiskey, or liquers, either alone or mixed?

Source releases
1
Available cycles
1999-2000
Dataset
Diet Behavior & Nutrition

No normalized longitudinal change evidence is available for this variable. Absence of events must not be interpreted as proof of cross-cycle equivalence.

Evidence provenance: Praxis automated inference from indexed CDC/NCHS evidence

Remove from studyVariable historyCDC source
DMARACE
Evidence developinglow confidence

Demographics

Logical Imputation Flag for Race Recode

Source releases
1
Available cycles
1999-2000
Dataset
Demographic Variables & Sample Weights

No normalized longitudinal change evidence is available for this variable. Absence of events must not be interpreted as proof of cross-cycle equivalence.

Evidence provenance: Praxis automated inference from indexed CDC/NCHS evidence

Remove from studyVariable historyCDC source
SDMVPSU
Compatible with harmonizationmoderate confidence

Demographics

Masked Variance Unit Pseudo-PSU variable for variance estimation

Source releases
12
Available cycles
1999-2000 · 2001-2002 · 2003-2004 · 2005-2006 · 2007-2008 · 2009-2010 · 2011-2012 · 2013-2014 · 2015-2016 · 2017-2018 · 2017-2020 · 2021-2023
Dataset
Demographic Variables & Sample Weights

No blocking method discontinuity is identified inside the selected cycle span, but one or more longitudinal changes require explicit harmonization conditions before pooling.

Evidence provenance: Praxis automated inference from indexed CDC/NCHS evidence

Evidence signals: source-derived · automated-difference

Relevant boundaries: 1999-2000 → 2001-2002 · 2001-2002 → 2003-2004 · 2003-2004 → 2005-2006 · 2005-2006 → 2007-2008 · 2007-2008 → 2009-2010 · 2009-2010 → 2011-2012 · 2011-2012 → 2013-2014 · 2013-2014 → 2015-2016 · 2015-2016 → 2017-2018 · 2017-2018 → 2017-2020 · 2017-2020 → 2021-2023

Praxis harmonization actions
  • Verify merge grain, companion variables, and component-specific analytic guidance for the new file context.
  • Select the weight appropriate to each cycle and subsample; when pooling cycles, construct the combined weight according to NHANES guidance rather than reusing a single-cycle weight unchanged.
  • Confirm that the construct and coding are unchanged before pooling or recoding.
Supporting CDC/NCHS evidence
Remove from studyVariable historyCDC source