Update FCSN-indicator-calculation.sps #199
Conversation
| *** In cases where 0 consumption of main group automatically codes subgroup values as missing, recoding to 0 consumption | ||
|
|
||
| RECODE FCSNPrMeatO FCSNPrMeatF FCSNPrFish FCSNPrEggs FCSNVegOrg FCSNVegGre FCSNFruiOrg (SYSMIS = 0). | ||
| EXECUTE. |
There was a problem hiding this comment.
Missing data imputation step: is it safe to assume every missing is 0?
| *** Harmonize Data Quality Guidance measures | ||
| *** Clean impossible values | ||
|
|
||
| *** Define labels | ||
|
|
||
| Variable labels | ||
| FCSNPrMeatF "Consumption in past 7 days: Flesh meat" | ||
| FCSNPrMeatO "Consumption in past 7 days: Organ meat" | ||
| FCSNPrFish "Consumption in past 7 days: Fish/shellfish" | ||
| FCSNPrEggs "Consumption in past 7 days: Eggs" | ||
| FCSNVegOrg "Consumption in past 7 days: Orange vegetables (vegetables rich in Vitamin A)" | ||
| FCSNVegGre "Consumption in past 7 days: Green leafy vegetables" | ||
| FCSNFruiOrg "Consumption in past 7 days: Orange fruits (Fruits rich in Vitamin A)". | ||
|
|
||
| *** Recode "n/a" values to 0 and change variable type to numeric | ||
|
|
||
| ALTER TYPE FCSNPrMeatF FCSNPrMeatO FCSNPrFish FCSNPrEggs FCSNVegOrg FCSNVegGre FCSNFruiOrg (a5). | ||
|
|
||
| RECODE FCSNPrMeatF FCSNPrMeatO FCSNPrFish FCSNPrEggs FCSNVegOrg FCSNVegGre FCSNFruiOrg | ||
| ('n/a'='0'). | ||
| RECODE FCSNPrMeatF FCSNPrMeatO FCSNPrFish FCSNPrEggs FCSNVegOrg FCSNVegGre FCSNFruiOrg (LOWEST THRU -1 = SYSMIS). | ||
| RECODE FCSNPrMeatF FCSNPrMeatO FCSNPrFish FCSNPrEggs FCSNVegOrg FCSNVegGre FCSNFruiOrg (8 THRU HIGHEST = SYSMIS). |
There was a problem hiding this comment.
Part of the logical data cleaning step
| *** Check flagged cases | ||
|
|
||
| FREQUENCIES VARIABLES = FCSN_flag_protein FCSN_flag_veg FCSN_flag_fruit | ||
| /ORDER = ANALYSIS. |
There was a problem hiding this comment.
Part of logical data cleaning steps
| VARIABLE LABELS Haem_iron_Cat 'Household consumption of haem iron'. | ||
| VALUE LABELS Haem_iron_Cat | ||
| 1 '0 time (never consumed)' | ||
| 2 '1-6 times (consumed sometimes)' |
There was a problem hiding this comment.
More of a question on the construction of the index, nothing to be addressed during review but for potential additional revision.
Being built on less categories (just 3), the value "7" is very high to reach. Are there any studies on the validation of this Iron intake approach and thresholds? (e.g.: sensitivity analysis)
Drafting a research question and finding a University Master/PhD student to perform robustness and sensitivity analysis on the choice of the thresholds this should be relatively straightforward.
2 participants
No description provided.