2017-06-28

[andkov]

At this moment 5 studies have submitted models with the following predictors:

> unique(ls_odds$octo$predictor) %>% as.character()
[1] "age"    "eduyrs" "cogact" "numfr"  "male"  

> unique(ls_odds$lasa$predictor) %>% as.character()
[1] "age"       "male"      "education" "cogpartic" "network"  

> unique(ls_odds$wh$predictor)%>% as.character()
[1] "age"      "male"     "tedtotyr" "leisure"  "netw"    

> unique(ls_odds$h70$predictor)%>% as.character()
[1] "age"     "sex"     "EDUYR"   "cogcomp" "socsize"

> unique(ls_odds$map$predictor)%>% as.character()
[1] "age"        "male"       "edu"        "cogact_old" "soc_net"  

Question

What are good labels for study-specific names of cognitive_activity and social_network_size? What should the names of these two harmonized variables be?

Tables

VIew tables in a Word file

[andkov]

View (A)

View (B)

View (C)

[andkov]

Sample size used in the model object

OCTO ---565
LASA ---2506
WH ---1285
H70 ---549
MAP ---1718

[andkov]

#Inegrated models from Johan

VIew A

VIew B

View C

[andkov]

whitehall has no 3-->2 transition, is this ok, @annierobi?

[annierobi]

Yes it's ok for Whitehall because she was unable to run the models without fixing some parameters.

…

On Tue, Jul 4, 2017 at 4:53 PM, Andriy V. Koval ***@***.***> wrote: whitehall has no 3-->2 transition, is this ok, @annierobi <https://github.com/annierobi>? [image: image] <https://user-images.githubusercontent.com/2608073/27842598-510d87e2-60d9-11e7-8535-1b60aa4bed3e.png> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#10 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKow3f-XgIMF0a1G1RlsD425FrFGHxhiks5sKqZkgaJpZM4OH1Xd> .

[andkov]

@annierobi , there re-made graphs and tables are ready.

WITHOUT Whitehall

• table
• View A: transition by predictor
• View C: study by transition

WITH Whitehall

• table
• View A: transition by predictor
• View C: study by transition