About stancemcgraw

stancemcgraw · ‎08-09-2020

So I am trying to this today with another character variable, comorbidities, and it's giving me the following error when I run the proc transpose step (proc sort data=long; by mitchid; run; proc transpose data=long out=wide prefix=comorbid_; by mitchID; ID comorbidities; IDLABEL comorbidities; var Count; run;) "ERROR: The ID value "comorbid_CVA" occurs twice in the same BY group. ERROR: The ID value "comorbid_DEMENTIA" occurs twice in the same BY group. ERROR: The ID value "comorbid_DEPENDENT_HLTH" occurs twice in the same BY group. ERROR: The ID value "comorbid_HYPERTENSION" occurs twice in the same BY group. ERROR: The ID value "comorbid_CHEMOTHERAPY" occurs twice in the same BY group. NOTE: The above message was for the following BY group: MITCHID= WARNING: 1647 observations omitted due to missing ID values. WARNING: 1 BY groups omitted due to earlier errors.

stancemcgraw · ‎08-07-2020

Hello, I have a huge dataset with a ton of different injury types, and am wondering if there is an easier way to create a different variable other than a bunch of of-then statements. I obviously don't need every type of injury to have its own category, and was hoping to group similar ones together, but it's taking so long. I attached the cause of injury list below. Here is the code I started typing: if cause_of_injury in ("FALL") then cause="FALL"; if cause_of_injury in ("FALL SL") then cause="FALL_SL"; if cause_of_injury in ("MVA") then cause="MVA"; if cause_of_injury in ("FALL HEIGHT") then cause="FALL_HEIGHT"; if cause_of_injury in ("FALL SL") then cause="FALL_SL"; if cause_of_injury in ("MCA") then cause="MCA"; if cause_of_injury in ("SKI", "SNOWB", "SMOB", "SLED", "SNOWB") then cause="SNOW"; if cause_of_injury in ("ASLT") then cause="ASLT"; if cause_of_injury in ("BIKE") then cause="BIKE"; There are sooo many categories. Any tips/ideas?

stancemcgraw · ‎08-07-2020

This worked too, thank you!

stancemcgraw · ‎08-06-2020

Thanks Reeza, The first part worked, but the array came back in my log as the following error: 74 data want; 475 set wide; 476 array _drugs(*) drugs_:; WARNING: Defining an array with zero elements. 477 do i=1 to dim(_drugs); 478 _drugs (i) = coalesce(_drugs(i), 0); 479 end; 480 481 **#calculate total drugs; 482 total_drugs = sum(of drugs_:); --- 71 ERROR 71-185: The SUM function call does not have enough arguments. 483 484 run; NOTE: The SAS System stopped processing this step because of errors. WARNING: The data set WORK.WANT may be incomplete. When this step was stopped there were 0 observations and 15 variables. WARNING: Data set WORK.WANT was not replaced because this step was stopped. NOTE: DATA statement used (Total process time): real time 0.06 seconds cpu time 0.04 seconds

stancemcgraw · ‎08-06-2020

Hi all, I'm trying to transpose my long data to wide data, but both variables are character variables. I'd like to count the number of drugs identified per study-ID. Some of the study-IDs also have more than 1 drug found. I also want to create a variable that is the total number of drugs per study-ID after its been transposed: Data have: Study-ID Drugs_found ID-1-RMC Opioid ID-2-RMC None ID-3-RMC Benzo ID-4-RMC Cocaine ID-4-RMC Opioid ID-5-RMC Meth Data want: Study-ID OPIOIDS METH NONE COCAINE BENZOS Total_drugs ID-1-RMC 1 0 0 0 0 1 ID-2-RMC 0 0 0 0 0 0 ID-3-RMC 0 0 0 0 1 1 ID-4-RMC 1 0 0 1 0 2 ID-5-RMC 0 1 0 0 0 1

stancemcgraw · ‎02-21-2020

Hello, I'm trying to overlay two scatterplots; one for bmi & days to transition and one for age & days to transition. I then want to draw a best fit line through them with results from my regression analysis. I've tried the following code proc template; define statgraph standard; begingraph; entrytitle "Model Transition days by Age and BMI"; layout overlay; scatterplot x=transition y=age; scatterplot x=transition y=bmi; endlayout; endgraph; end; run; proc sgrender data=sashelp.class template=standard; run; It will not output the graph because it says I am missing arguments in the log. Could someone please help p me out?

stancemcgraw · ‎12-10-2019

Does anyone know the code to do a chart like this, in SAS 9.4? Where the survey questions are on the y axis and the survey scores are on the x-axis, divided up in the legend between pre and post, plus standard deviations. See attached photo example

stancemcgraw · ‎11-06-2019

Would you know the code to reproduce what I did above for sgplot?

stancemcgraw · ‎11-06-2019

Hi all, I have the following code that correctly produced three histograms by group. However, I would 1) like to change the color of the bars (fill black) 2) change the x axis label from Npo_mean to Hours NPO:. 3) Change and clarify the Y-axis group labels The graph output is attached below as well proc univariate data=tube.survival2; class npogroup ; var Hours; /* computes descriptive statisitcs */ histogram Hours / odstitle="Distribution of NPO by Fasting Category" vscale=percent vaxis=0 to 60 by 10 normal nrows=3 midpoints=(0 to 22 by 2) ; ods select histogram; /* display on the histograms */ INSET N = 'Number of patients' MEDIAN MEAN /position=ne ; run;

stancemcgraw · ‎08-06-2019

I'm trying to create a variable that counts when a procedure says "CT", "XRAY", "MRI" and call it Imaging and then so on and so forth for other types of variable groups I want. Is there a scan or count or find function that would work in this scenario? data have: studynum procs 1 xray 1 leg xray 2 CT 2 head CT 3 MRI 3 Ortho data want; studynum procs imaging 1 xray 1 1 leg xray 1 2 CT 1 2 head CT 1 3 MRI 1 3 Ortho 0

stancemcgraw · ‎07-23-2019

I would like to categorize and count procedure type by studynum. What is the easiest way with the least amount of text? Data have: procedures studynum xray 5 xray 5 orif 5 irrigation 5 CT 6 ETT 6 central line 6 data want: studynum xray orif irrigation CT ETT Central line 5 2 1 1 0 0 0 6 1 0 0 1 1 1 I also want to group them. So CT and Xray would be Imaging=1. And irrigation would be Wound/Skin=1;

stancemcgraw · ‎06-18-2019

What does the LAG function do?

stancemcgraw · ‎06-18-2019

Hi all, I am trying to calculate the time from the last tube feed (tube_stopped) to the time of surgery. Each patient has multiple tube stop times and also multiple surgeries (date_procedure). I want to create a new variable that calculates this called "Time from last tube feed to procedure" in hours.... Data have: ID tube_stopped date_procedure 7 12Jul17:02:00:00 13JUL17:13:00:00 7 01AUG17:15:00:00 01AUG17:22:00:00 7 20JUL17:07:35:00 20JUL17:22:00:00 1 29JUL16:01:00:00 29JUL16:23:00:00 1 12AUG16:16:00:00 12AUG16:22:00:00 Data want: ID tube_stopped date_procedure time from last tube feed to procedure 7 01AUG17:15:00:00 01AUG17:22:00:00 2.4 1 12AUG16:16:00:00 12AUG16:22:00:00 5.7

stancemcgraw · ‎04-18-2019

I would like to create SAS code that calculates the total days between cases of debridement by studynum. I don't know how to do that.

stancemcgraw · ‎04-18-2019

DI and Debridement will be counted as the same thing. So yes. I just put 5.3 there is an example. I don't know the actual number

Online Status	Offline
Date Last Visited	‎12-21-2023 05:08 PM

Character dates and times to numeric dates and times

Code out wide data with an array

Coding data wide to long

Re: Character dates

Re: Character dates

Character dates

Trying to find difference between first and last date of same subject

Re: Identify and remove hidden characters in character variable

Re: Identify and remove hidden characters in character variable

Re: Identify and remove hidden characters in character variable

Re: Convert character hours and minutes to numeric days

Re: Formatting 24 hr character time to 24hr numeric time

Re: Change character date in format $13. to ddmmyy10. numeric

Re: Create new dummy variables from multiple columns of character vari...

Re: Scan or Find or Count option

Code out wide data with an array

Re: Transpose character data wide

Characterizing using IF-THEN STATEMENTS

Re: Transpose character data wide

Re: Transpose character data wide

Transpose character data wide

Overlay two scatterplots and draw line of best fit

Bar charts

Re: Histogram labels and colors

Histogram labels and colors

Scan or Find or Count option

Group by class

Re: Proc transpose

Proc transpose

Re: Transposing and Calculating

Re: Transposing and Calculating