Alcohol

[caitlink12]

Alcohol variables ALCDTTM, ALWDWKY, binge drinker and it's dependents (ALW_2A1, ALW_2A2, ALW_2A3, ALW_2A4, ALW_2A5, ALW_2A6, ALW_2A7) were updated for ICES survey cycles in variable and variable_details sheets.

[DougManuel]

Review: PR #166 (Alcohol)

Reviewed the _i → _m conversion for 11 alcohol variables. L6 integration testing passed for all 9 PUMF cycles — no step changes at era boundaries.

Full review details in CEP-011.

P0: ALW_2A1 row 1 — wrong source variables (fixed)

ALW_2A1 row 1 (recStart=[0,50]) had ALW_1's source variables instead of ALW_2A1's. Copy-paste error.

Was (wrong):

cchs2001_p::ALCA_5, cchs2003_p::ALCC_5, cchs2005_p::ALCE_5,
cchs2015_2016_p::ALW_005, cchs2017_2018_p::ALW_005, [ALW_1]

Now (corrected, matching rows 2-4):

cchs2001_p::ALCA_5A1, cchs2003_p::ALCC_5A1, cchs2005_p::ALCE_5A1,
cchs2015_2016_p::ALW_010, cchs2017_2018_p::ALW_010,
cchs2001_m::ALCA_5A1, cchs2003_m::ALCC_5A1, cchs2005_m::ALCE_5A1,
cchs2015_2016_m::ALW_010, cchs2017_2018_m::ALW_010, [ALW_2A1]

The _5 variables are "any alcohol past week" (yes/no); the _5A1 variables are "number of drinks on Sunday" (continuous). At runtime this would have silently produced wrong values.

P1: _s databases → _m (fixed)

Converted cchs2009_s/cchs2010_s/cchs2012_s to cchs2009_m/cchs2010_m/cchs2012_m in databaseStart for all 11 alcohol variables (48 rows). variables.csv already had _m.

Both fixes are in commit 30667cf.

[rafdoodle]

Alcohol variables — _m database update

Completed (11 variables)

Updated databaseStart and variableStart in both variables.csv and variable_details.csv:

• Replaced legacy _s database entries with full _m equivalents
• Added missing singular-year _m entries (cchs2009_m, cchs2010_m, cchs2011_2012_m, cchs2012_m) in correct chronological order
• Added explicit _m source variable mappings in variableStart
• Fixed miscellaneous formatting issues (trailing spaces, double spaces, incorrect copy-row variableStart)

Variables:

• ALCDTTM — two-group structure preserved (pre-/post-2007)
• ALW_1
• ALW_2A1 — also fixed copy-row variableStart; removed trailing space
• ALW_2A2–ALW_2A7
• ALWDWKY — fixed space after :: in variableStart
• binge_drinker

Still needs _m updating

• ALC_005 — no _m entries; only covers 2001–2007/2008 _p
• ALC_1 — has only cchs2009_m, cchs2010_m, cchs2012_m; needs full _m set + explicit variableStart mappings
• ALCDTYP — no _m entries; only covers 2001–2007/2008 _p
• ALCDTYP_A — typo 2011_2012_m (missing cchs prefix); missing cchs2009_m, cchs2010_m, cchs2012_m singular years
• ALWDDLY — has only cchs2009_m, cchs2010_m, cchs2012_m; needs full _m set + explicit variableStart mappings; double space in databaseStart; trailing spaces in variableStart
• ALWDVLTR_der — has only cchs2009_m, cchs2010_m, cchs2012_m; needs full _m set; double space in databaseStart
• ALWDVSTR_der — has only cchs2009_m, cchs2010_m, cchs2012_m; needs full _m set; double space in databaseStart
• low_drink_score — has only cchs2009_m, cchs2010_m, cchs2012_m; needs full _m set + explicit variableStart mappings
• low_drink_score1 — no _m entries; only covers select _p years

[rafdoodle]

All alcohol variables look good now.

@yulric @DougManuel low_drink_score_fun() and low_drink_score_fun1() are not in R/alcohol.R, bur rather in R/legacy/alcohol-legacy.R. Does that mean they are no longer being used and thus low_drink_score and low_drink_score1 should be deleted? Thanks.