topic Re: How to switch from short to long format using data steps in SAS Programming

How to switch from short to long format using data steps

Stacy — Sat, 12 Jan 2019 19:44:36 GMT

https://s3.us-east-1.amazonaws.com/blackboard.learn.xythos.prod/5954eb74c7df4/3105111?response-content-disposition=inline%3B%20filename%2A%3DUTF-8%27%27epilepsy.txt&response-content-type=text%2Fplain&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20190112T193338Z&X-Amz-SignedHeaders=host&X-Amz-Expires=21600&X-Amz-Credential=AKIAIL7WQYDOOHAZJGWQ%2F20190112%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=2640da794352971ad9807a6932c9f455c938efcb1390a0e5f12c5715cb6752e6

Hi,

Above is the link of the file that I need to switch from short to long format using data steps.

I attached the project questions, I am very new to programming and have no idea on how to continue.

I am hoping that someone here could please help me out.

First,

I imported the text file to SAS which I did successfully

Now I have to change the format from short to long

I don’t know how to approach this because SAS says the file has only one variable while there are about 6 or more columns.

Re: How to switch from short to long format using data steps

Reeza — Sat, 12 Jan 2019 21:36:18 GMT

That means you did not import the data correctly. Please post the code and log from the import step. It will say the number of variables and observations which is one indication that the file was read correctly.

@Stacy wrote:

https://s3.us-east-1.amazonaws.com/blackboard.learn.xythos.prod/5954eb74c7df4/3105111?response-content-disposition=inline%3B%20filename%2A%3DUTF-8%27%27epilepsy.txt&response-content-type=text%2Fplain&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20190112T193338Z&X-Amz-SignedHeaders=host&X-Amz-Expires=21600&X-Amz-Credential=AKIAIL7WQYDOOHAZJGWQ%2F20190112%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=2640da794352971ad9807a6932c9f455c938efcb1390a0e5f12c5715cb6752e6

Hi,

Above is the link of the file that I need to switch from short to long format using data steps.

I attached the project questions, I am very new to programming and have no idea on how to continue.

I am hoping that someone here could please help me out.

First,

I imported the text file to SAS which I did successfully

Now I have to change the format from short to long

I don’t know how to approach this because SAS says the file has only one variable while there are about 6 or more columns.

Re: How to switch from short to long format using data steps

Stacy — Sat, 12 Jan 2019 22:03:46 GMT

Libname sasuser 'C:\Users\Ifeoma\Downloads';
Data sasuser.epilepsy;
  SET epilepsy;
  input ID $ Treatment Age ESC TSC1 TSC2 TSC3 TSC4 $;
  datalines;
           1     0      31      11      5     3     3     3
           2     0      30      11      3     5     3     3
           3     0      25       6      2     4     0     5
           4     0      36       8      4     4     1     4
           5     0      22      66      7    18     9    21
           6     0      29      27      5     2     8     7
           7     0      31      12      6     4     0     2
           8     0      36      52     40    20    23    12
           9     0      37      23      5     6     6     5
          10     0      28      10     14    13     6     0
          11     0      36      52     26    12     6    22
          12     0      24      33     12     6     8     5
          13     0      28      18      4     4     6     2
          14     0      36      42      7     9    12    14
          15     0      26      87     16    24    10     9
          16     0      26      50     11     0     0     5
          17     0      28      18      0     0     3     3
          18     0      31     111     37    29    28    29
          19     0      32      18      3     5     2     5
          20     0      21      20      3     0     6     7
          21     0      29      12      3     4     3     4
          22     0      21       9      3     4     3     4
          23     0      32      17      2     3     3     5
          24     0      25      28      8    12     2     8
          25     0      30      55     18    24    76    25
          26     0      40       9      2     1     2     1
          27     0      19      10      3     1     4     2
          28     0      22      47     13    15    13    12
          29     1      18      76     11    14     9     8
          30     1      32      38      8     7     9     4
          31     1      20      19      0     4     3     0
          32     1      20      10      3     6     1     3
          33     1      18      19      2     6     7     4
          34     1      24      24      4     3     1     3
          35     1      30      31     22    17    19    16
          36     1      35      14      5     4     7     4
          37     1      57      11      2     4     0     4
          38     1      20      67      3     7     7     7
          39     1      22      41      4    18     2     5
          40     1      28       7      2     1     1     0
          41     1      23      22      0     2     4     0
          42     1      40      13      5     4     0     3
          43     1      43      46     11    14    25    15
          44     1      21      36     10     5     3     8
          45     1      35      38     19     7     6     7
          46     1      25       7      1     1     2     4
          47     1      26      36      6    10     8     8
          48     1      25      11      2     1     0     0
          49     1      22     151    102    65    72    63
          50     1      32      22      4     3     2     4
          51     1      25      42      8     6     5     7
          52     1      35      32      1     3     1     5
          53     1      21      56     18    11    28    13
          54     1      41      24      6     3     4     0
          55     1      32      16      3     5     4     3
          56     1      26      22      1    23    19     8
          57     1      21      25      2     3     0     1
          58     1      36      13      0     0     0     0
          59     1      37      12      1     4     3     2
;
run;
Data epilepsy1;
  Set epilepsy;
  Week = 1;
  Seizurecount = TSC1;
  drop TSC1 TSC2 TSC3 TSC4;
run;
Data epilepsy2;
  Set epilepsy;
  Week = 2;
  Seizurecount = TSC2;
  drop TSC1 TSC2 TSC3 TSC4;
run;
Data epilepsy3;
  Set epilepsy;
  Week = 3;
  Seizurecount = TSC3;
  drop TSC1 TSC2 TSC3 TSC4;
run;
Data epilepsy4;
  Set epilepsy;
  Week = 4;
  Seizurecount = TSC4;
  drop TSC1 TSC2 TSC3 TSC4;
run;
Data sasuser.epilepsylong;
  Set epilepsy1 epilepsy2 epilepsy3 epilepsy4;
run;

After running the code, I got the following errors;

NOTE: Variable TSC1 is uninitialized.
WARNING: The variable TSC2 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC3 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC4 in the DROP, KEEP, or RENAME list has never been referenced.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY.
NOTE: The data set WORK.EPILEPSY1 has 94 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           0.00 seconds
      cpu time            0.01 seconds


2243  Data epilepsy2;
2244    Set epilepsy;
2245    Week = 2;
2246    Seizurecount = TSC2;
2247    drop TSC1 TSC2 TSC3 TSC4;
2248  run;

NOTE: Variable TSC2 is uninitialized.
WARNING: The variable TSC1 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC3 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC4 in the DROP, KEEP, or RENAME list has never been referenced.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY.
NOTE: The data set WORK.EPILEPSY2 has 94 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           0.00 seconds
      cpu time            0.00 seconds


2249  Data epilepsy3;
2250    Set epilepsy;
2251    Week = 3;
2252    Seizurecount = TSC3;
2253    drop TSC1 TSC2 TSC3 TSC4;
2254  run;

NOTE: Variable TSC3 is uninitialized.
WARNING: The variable TSC1 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC2 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC4 in the DROP, KEEP, or RENAME list has never been referenced.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY.
NOTE: The data set WORK.EPILEPSY3 has 94 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           0.01 seconds
      cpu time            0.01 seconds


2255  Data epilepsy4;
2256    Set epilepsy;
2257    Week = 4;
2258    Seizurecount = TSC4;
2259    drop TSC1 TSC2 TSC3 TSC4;
2260  run;

NOTE: Variable TSC4 is uninitialized.
WARNING: The variable TSC1 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC2 in the DROP, KEEP, or RENAME list has never been referenced.
WARNING: The variable TSC3 in the DROP, KEEP, or RENAME list has never been referenced.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY.
NOTE: The data set WORK.EPILEPSY4 has 94 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           0.01 seconds
      cpu time            0.01 seconds


2261  Data sasuser.epilepsylong;
2262    Set epilepsy1 epilepsy2 epilepsy3 epilepsy4;
2263  run;

NOTE: There were 94 observations read from the data set WORK.EPILEPSY1.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY2.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY3.
NOTE: There were 94 observations read from the data set WORK.EPILEPSY4.
NOTE: The data set SASUSER.EPILEPSYLONG has 376 observations and 3 variables.
NOTE: DATA statement used (Total process time):
      real time           0.01 seconds
      cpu time            0.01 seconds

Re: How to switch from short to long format using data steps

Reeza — Sat, 12 Jan 2019 22:19:12 GMT

Look at the name of the data set you saved the intial data to and the data set you’re trying to use later. They’re not the same, so you’re not trying to ise the file you imported.

I’ll give you one tip to avoid these type of errors - comment your code, every line, with what you think it is doing or should be doing.

Re: How to switch from short to long format using data steps

Stacy — Sat, 12 Jan 2019 22:48:24 GMT

I dont see what youre talking about. The data sets I used are the same.

Re: How to switch from short to long format using data steps

Patrick — Sun, 13 Jan 2019 01:30:06 GMT

@Stacy wrote:

I dont see what youre talking about. The data sets I used are the same.

No they aren't.

Tables are referenced using <libref>.<table name>. If you only use the table name like in "set epilepsy" then SAS defaults to WORK as your libref.

Now look at your code:

Libname sasuser 'C:\Users\Ifeoma\Downloads';
Data sasuser.epilepsy;
  SET epilepsy;
  input ID $ Treatment Age ESC TSC1 TSC2 TSC3 TSC4 $;
  datalines;

You are creating a table under libref SASUSER but then later in your code you try to use the table under WORK (as you're only using a one level name).

Also: What's this combination of SET and INPUT statement about? That's most likely not going to work as you'd expect.

You also might want to remove the $ at the end of your INPUT line. Why would you want to only read TSC4 as character.

And last but not least: Not sure why you need to replicate your data just with a different number for Week - but if really necessary then code as below would be more efficient

data epilepsylong(drop=TSC:);
  set epilepsy;
  array TSC {*} TSC1-TSC4;
  do Week=1 to 4;
    Seizurecount=TSC[Week];
    output;
  end;
run;

Re: How to switch from short to long format using data steps

Tom — Sun, 13 Jan 2019 01:30:34 GMT

Your first step doesn't make any sense.

Data sasuser.epilepsy;
  SET epilepsy;
  input ID $ Treatment Age ESC TSC1 TSC2 TSC3 TSC4 $;
  datalines;

Why do you have both a SET statement and an INPUT statement.

That would mean that you are assuming that the existing dataset EPILEPSY has the same number of observations as the data that will be read from the datalines AND that the observations are in the exact same order so that it makes sense to add new variables (or overwrite the existing variables) with the data from the datalines.

Re: How to switch from short to long format using data steps

Stacy — Sun, 13 Jan 2019 01:55:18 GMT

I used both a set and input statement because in the file theres no column specofying the variable.

;Likethis;

Patient ID, Treatment (0=Placebo, 1=Progabide), Age, 
Baseline 8 week seizure count, First 2 week seizure count,
Second 2 week seizure count, Third 2 week seizure count,
Fourth 2 week seizure count.


           1     0      31      11      5     3     3     3
           2     0      30      11      3     5     3     3
           3     0      25       6      2     4     0     5
           4     0      36       8      4     4     1     4
           5     0      22      66      7    18     9    21
           6     0      29      27      5     2     8     7
           7     0      31      12      6     4     0     2
           8     0      36      52     40    20    23    12
           9     0      37      23      5     6     6     5
          10     0      28      10     14    13     6     0
          11     0      36      52     26    12     6    22
          12     0      24      33     12     6     8     5
          13     0      28      18      4     4     6     2
          14     0      36      42      7     9    12    14
          15     0      26      87     16    24    10     9
          16     0      26      50     11     0     0     5
          17     0      28      18      0     0     3     3
          18     0      31     111     37    29    28    29
          19     0      32      18      3     5     2     5
          20     0      21      20      3     0     6     7
          21     0      29      12      3     4     3     4
          22     0      21       9      3     4     3     4

The numbers dont have a heading, so I thought by using input I could give the columns a heading before converting from short format to long format.

Re: How to switch from short to long format using data steps

Stacy — Sun, 13 Jan 2019 01:59:52 GMT

Also, the set recorded as 94 observations with one variable, so I used the input statement to create variables for each column.

Re: How to switch from short to long format using data steps

Patrick — Sun, 13 Jan 2019 02:27:09 GMT

SET is used to read data from a table, INPUT is used to read data from an external file (or inline via DATALINES).

The @ used at the end of an INPUT statement keeps the input pointer on the same line. This allows you to read the data and directly output it into a long format.

data epilepsy;
  infile datalines truncover;
  input ID $ Treatment Age ESC @;
  do week=1 to 4;
    input Seizurecount @;
    output;
  end;
  datalines;
1     0      31      11      5     3     3     3
2     0      30      11      3     5     3     3
3     0      25       6      2     4     0     5
4     0      36       8      4     4     1     4
5     0      22      66      7    18     9    21
6     0      29      27      5     2     8     7
7     0      31      12      6     4     0     2
8     0      36      52     40    20    23    12
9     0      37      23      5     6     6     5
10     0      28      10     14    13     6     0
11     0      36      52     26    12     6    22
12     0      24      33     12     6     8     5
13     0      28      18      4     4     6     2
14     0      36      42      7     9    12    14
15     0      26      87     16    24    10     9
16     0      26      50     11     0     0     5
17     0      28      18      0     0     3     3
18     0      31     111     37    29    28    29
19     0      32      18      3     5     2     5
20     0      21      20      3     0     6     7
21     0      29      12      3     4     3     4
22     0      21       9      3     4     3     4
23     0      32      17      2     3     3     5
24     0      25      28      8    12     2     8
25     0      30      55     18    24    76    25
26     0      40       9      2     1     2     1
27     0      19      10      3     1     4     2
28     0      22      47     13    15    13    12
29     1      18      76     11    14     9     8
30     1      32      38      8     7     9     4
31     1      20      19      0     4     3     0
32     1      20      10      3     6     1     3
33     1      18      19      2     6     7     4
34     1      24      24      4     3     1     3
35     1      30      31     22    17    19    16
36     1      35      14      5     4     7     4
37     1      57      11      2     4     0     4
38     1      20      67      3     7     7     7
39     1      22      41      4    18     2     5
40     1      28       7      2     1     1     0
41     1      23      22      0     2     4     0
42     1      40      13      5     4     0     3
43     1      43      46     11    14    25    15
44     1      21      36     10     5     3     8
45     1      35      38     19     7     6     7
46     1      25       7      1     1     2     4
47     1      26      36      6    10     8     8
48     1      25      11      2     1     0     0
49     1      22     151    102    65    72    63
50     1      32      22      4     3     2     4
51     1      25      42      8     6     5     7
52     1      35      32      1     3     1     5
53     1      21      56     18    11    28    13
54     1      41      24      6     3     4     0
55     1      32      16      3     5     4     3
56     1      26      22      1    23    19     8
57     1      21      25      2     3     0     1
58     1      36      13      0     0     0     0
59     1      37      12      1     4     3     2
;
run;

If your actual data is stored in an external file then eventually post this file (or some of the first lines of the file) here as attachment so that we can show how to read such a file (uses of a FILENAME statement).

Re: How to switch from short to long format using data steps

Stacy — Sun, 13 Jan 2019 03:18:25 GMT

Hi,

Thanks for the reply. I imported the file to SAS. I attached the file.

How do I convert it to long format without using datalines

Re: How to switch from short to long format using data steps

Stacy — Sun, 13 Jan 2019 03:22:07 GMT

Hi Patrick,

The code worked thanks.

Thank you

Re: How to switch from short to long format using data steps

Patrick — Sun, 13 Jan 2019 03:39:00 GMT

Good to hear and thanks for posting your external file.

Given that it's not a lot of data I believe easiest for you is to go with the inline data (datalines) approach which works already for you.

Re: How to switch from short to long format using data steps

Stacy — Sun, 13 Jan 2019 03:47:55 GMT

Yes that is true and thanks.

Please, how can I go in converting the ESC column and Seizure count column into weekly rates

Re: How to switch from short to long format using data steps

Patrick — Sun, 13 Jan 2019 05:03:40 GMT

@Stacy wrote:

Yes that is true and thanks.

Please, how can I go in converting the ESC column and Seizure count column into weekly rates

I have no idea what these "weekly rates" are so you would have to describe the logic in detail and provide examples based on the sample data posted.

Re: How to switch from short to long format using data steps

Reeza — Mon, 14 Jan 2019 02:47:12 GMT

@Stacy Please go through and mark some of your previous questions as solved.