Friedman in stata vs in Prism

[Zainab Mahmood]

when I do Friedman in stata, it gives me different P value than when I did it in Graphpad prism. In Prism, I delete manually the rows with missing values. I have three measures over time for each individual. in stata I just run the command friedman var1 var2 var3. from both the significance, non significance persist but with different numbers (not exact P valuses). Any explanation? or I might be doing something wrong.

[Rich Goldstein]

I am the original author of the -friedman- command but I never use it (it was written because one particular client at that time, many years ago, wanted it); however, I know nothing about Graphpad prism so I don't know what they are doing; if you post an example of your data, using -dataex- and posting within CODE blocks (as per the FAQ), I may be able to help re: what the -friedman- command is doing

[Zainab Mahmood]

Thanks for your response. Glad to speak to the original author
This is my data after I made it wide (in terms of visits) but still long in terms of group (treated vs placebo). My other question please any way to tell stata to do friedman for treated and for placebo separately? as I tried by or if, none has worked. I did two separate sheet for each manually.
dataex ipiratioauc1 ipiratioauc2 ipiratioauc3 group

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. To install: ssc install dataex
clear
input double(ipiratioauc1 ipiratioauc2 ipiratioauc3) long group
.2330269912647059 .2451710845064207 .1450165625092701 1
.2025895901843955 .1816664950690817 .1446290509307822 1
.1334045540475425 .1478527417681296                 . 1
.1803098106073392 .2960839776519851 .2962146655762097 1
.3840034890859194 .3605358429055334 .3245378121412726 1
                .                 . .4421104173244128 1
.2165484537586932        .016996982 .1512413092868024 1
.1122716819553272 .3808535388134742 .1062763059639889 1
.1537766172195993 .1352683750778654 .0674639223427435 1
.4985534368736553 1.107074055662806 .3563916074063879 1
                . .3850331379183735 .1533277929333614 1
.4280923938925788 .5371577241209082 .2671287250631313 1
                .                 . .2769027704682041 1
                . .7322294843822921 .7239095766804513 1
.1299902401889971 .1214012216712829 .0741180905422245 1
       .115436838        .079804181         .11126806 1
                . .8935047708333334 1.502373416555219 1
                .                 . .1896127803507767 1
 .208616083058608 .1204280313852814 .3031954041192885 1
.0838132043379123 .1013362406365611 .1454422451914388 1
.1729400386375942 .1211358880822747 .3009137006408532 1
.3409979639775547 .4176234471500722  .154126952268106 1
.2714260464110263 .5744495070087033                 . 1
 .228836347565508 .2106821916942338 .1024385390356601 1
.4577974003988178 .2571346177876655 .1443217481560892 1
                . .2050237882553859  .285036862063371 2
.3579258836673384 .2745258705947867 .1974079820546885 2
.9780011666666666 .7282311920426063                 . 2
                . .8841236709956709 1.161059750738733 2
  .18487907766742                 .                 . 2
                . .1976001691897313 .3161802743853649 2
.5306283146856178                 . .2830309469397261 2
       .733131149        .031197586                 . 2
.2666496999933066 .1853639275331153 .5470211903809692 2
                . .3703660713084189 .8170944556288171 2
.4933354121239187 .6248593946306135 .1658427598743256 2
                .                 . .3509823211129148 2
                .                 . .2476537359325336 2
.2053817702629536                 .                 . 2
.4005153054617118 .4232648534798535 .0919818739177489 2
.1167602736064293 .0230625291997884  .066829338973725 2
.5416112094414557                 . .2279348745421245 2
.2751565015609391 .2561298615032311   .19092600464145 2
                . .1722850359631977 .0622006012421975 2
                .                 . .1432569161852058 2
1.672809111560009                 . .2916368874482401 2
.2462737274985221 .3063995067833407                 . 2
                . .7384097427869202 .4499512707605788 2
.5302095212722586 .2365258182362831  .495987531474973 2
.2253974050038832 .2809892368313454 .5131283830673413 2
.2566043439235109 .2083182404127154 .1357947677975196 2
.4021764182183525 .1581732258623275  .460590113328449 2
.2457732078562724 .3177858205960291 .1159365758749667 2
.2379704571118768 .3269683664772727 .2513458175391816 2
.7140170921717173 .7015430181818182 .2208647294001882 2
.1003674147321429 .3372831506944445 .1346491703296703 2
 .486456713762383 .4511184596492807 .7028756906512605 2
.3215618464457904 .4857079248062115  .956133200652967 2
.9077072940211962 .8707027824044614 .2393762771256586 2
 .252582438085562 .2095642651407327 .1606658377329796 2
.2047018646404274 .1567282271557272 .0618595302903753 2
       .045603448        .013103448  .728554823761329 2
                .        .346658709 .0635292477300579 2
.5832751513401039 .2252925709067057 .2353279536886975 2
.1164260840428785 .3289489641751926 .1299653560516747 2
3.479824666666667 1.823899575757576                 . 2
                .                 .                 . 2
1.702252030375467 .5495861672150992 .2560568488773483 2
.2168245364222466 .2003923270318129 .1096176134673863 2
                .        .005717011 .2142233056690517 2
.3759020778528802                 .                 . 2
                . .5385350827353729 .2413912873477541 2
end
label values group group1
label def group1 1 "Saline", modify
label def group1 2 "Treated", modify

the data has Patient ID as well

[Rich Goldstein]

here is the overall result I have:

. friedman ip*

Friedman =  74.7085
Kendall =    0.6226
P-value =    0.0007

and here is an example for one group (the other should be obvious):

. friedman ip* if group==1

Friedman =  30.4575
Kendall =    0.6345
P-value =    0.0158

[Zainab Mahmood]

Thank you very much. it is different from what I got, unfortunately. Can you explain the code please so I can reproduce what you did. This data is only example even it is the full set for those 3 variables. within the same sheet I have more other variables named ipi01 ipi02 ipi03, ipiratio01 ipiratio02 ipiratio03..etc
what ip* means?

[Zainab Mahmood]

Sorry I did it for group2 , I did it now for group1 it is exactly the same... Thanks a lot much appreciated. I used freidman ipiratioauc1 ipiratioauc2 ipiratioauc3 if group==1 and it worked!

[Rich Goldstein]

you're welcome and good luck