SEL-Columbia / nmis_R_scripts

R scripts for NMIS
Apache License 2.0
1 stars 3 forks source link

Inconsistency in the definition of pupil_toilet_ratio_facility #48

Closed zmyao88 closed 11 years ago

zmyao88 commented 11 years ago

in 661: ed$pupil_toilet_ratio_facility <- e$num_students_total_gender.num_students_total / (e$num_toilet.num_toilet_boy + e$num_toilet.num_toilet_girl + e$num_toilet.num_toilet_both)

in 113:

ity <- (ifelse(e_113$flush_toilet_drain_to == "improved", rowSums(cbind(e_113$flush_toilet_number, as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number)), na.rm=T), rowSums(cbind(as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number), ed$education_improved_sanitation), na.rm=T)) / e_113$num_students_total )

which is totally upside-down.

To do: Talk to sector team and figure out the right definition.

prabhasp commented 11 years ago

Its "pupil to toilet" so pupils should be in the numerator.

Yay normalization and finding bugs!

On Tue, Sep 24, 2013 at 4:44 PM, Zaiming Yao notifications@github.comwrote:

in 661: ed$pupil_toilet_ratio_facility <- e$num_students_total_gender.num_students_total / (e$num_toilet.num_toilet_boy + e$num_toilet.num_toilet_girl + e$num_toilet.num_toilet_both)

in 113:

ity <- (ifelse(e_113$flush_toilet_drain_to == "improved", rowSums(cbind(e_113$flush_toilet_number, as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number)), na.rm=T), rowSums(cbind(as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number), ed$education_improved_sanitation), na.rm=T)) / e_113$num_students_total )

which is totally upside-down.

To do: Talk to sector team and figure out the right definition.

— Reply to this email directly or view it on GitHubhttps://github.com/modilabs/nmis_R_scripts/issues/48 .

Prabhas Pokharel http://prabhasp.com US mobile: +1 347 948 7654 twitter/skype/facebook/whatever: prabhasp

schafik commented 11 years ago

i think we should run this by education just to be 100% sure. i remember some ratio being calculated as the opposite of its titular definition....

On Tue, Sep 24, 2013 at 4:46 PM, prabhasp notifications@github.com wrote:

Its "pupil to toilet" so pupils should be in the numerator.

Yay normalization and finding bugs!

On Tue, Sep 24, 2013 at 4:44 PM, Zaiming Yao notifications@github.comwrote:

in 661: ed$pupil_toilet_ratio_facility <- e$num_students_total_gender.num_students_total / (e$num_toilet.num_toilet_boy + e$num_toilet.num_toilet_girl + e$num_toilet.num_toilet_both)

in 113:

ity <- (ifelse(e_113$flush_toilet_drain_to == "improved", rowSums(cbind(e_113$flush_toilet_number, as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number)), na.rm=T), rowSums(cbind(as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number), ed$education_improved_sanitation), na.rm=T)) / e_113$num_students_total )

which is totally upside-down.

To do: Talk to sector team and figure out the right definition.

— Reply to this email directly or view it on GitHub< https://github.com/modilabs/nmis_R_scripts/issues/48> .

Prabhas Pokharel http://prabhasp.com US mobile: +1 347 948 7654 twitter/skype/facebook/whatever: prabhasp

— Reply to this email directly or view it on GitHubhttps://github.com/modilabs/nmis_R_scripts/issues/48#issuecomment-25040956 .

prabhasp commented 11 years ago

feel free to check, but we should be consistent in indicator titles + formulas. if toilets is on top, its toilet to pupil ratio (or some other wording that indicates that it is not a ratio of pupils to toilets).

On Tue, Sep 24, 2013 at 4:52 PM, schafik notifications@github.com wrote:

i think we should run this by education just to be 100% sure. i remember some ratio being calculated as the opposite of its titular definition....

On Tue, Sep 24, 2013 at 4:46 PM, prabhasp notifications@github.com wrote:

Its "pupil to toilet" so pupils should be in the numerator.

Yay normalization and finding bugs!

On Tue, Sep 24, 2013 at 4:44 PM, Zaiming Yao notifications@github.comwrote:

in 661: ed$pupil_toilet_ratio_facility <- e$num_students_total_gender.num_students_total / (e$num_toilet.num_toilet_boy + e$num_toilet.num_toilet_girl + e$num_toilet.num_toilet_both)

in 113:

ity <- (ifelse(e_113$flush_toilet_drain_to == "improved", rowSums(cbind(e_113$flush_toilet_number, as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number)), na.rm=T), rowSums(cbind(as.numeric(e_113$vip_latrine_number), as.numeric(e_113$slab_pit_latrine_number), ed$education_improved_sanitation), na.rm=T)) / e_113$num_students_total )

which is totally upside-down.

To do: Talk to sector team and figure out the right definition.

— Reply to this email directly or view it on GitHub< https://github.com/modilabs/nmis_R_scripts/issues/48> .

Prabhas Pokharel http://prabhasp.com US mobile: +1 347 948 7654 twitter/skype/facebook/whatever: prabhasp

— Reply to this email directly or view it on GitHub< https://github.com/modilabs/nmis_R_scripts/issues/48#issuecomment-25040956>

.

— Reply to this email directly or view it on GitHubhttps://github.com/modilabs/nmis_R_scripts/issues/48#issuecomment-25041475 .

Prabhas Pokharel http://prabhasp.com US mobile: +1 347 948 7654 twitter/skype/facebook/whatever: prabhasp

zmyao88 commented 11 years ago

fixed in 45_normalization close