This module illustrates how to reshape data
files in SPSS without using the varstocases command. These examples take wide data files and reshape them into
long form. These
show common examples of reshaping data but do not exhaustively demonstrate the
different kinds of data reshaping that you could encounter.
Example #1: One variable
Consider the file containing the kids and their heights at
one year of age
(ht1) and at two years of age (ht2).
get file 'c:kidshtwt.sav'. list famid birth ht1 ht2.FAMID BIRTH HT1 HT2 1.00 1.00 2.80 3.40 1.00 2.00 2.90 3.80 1.00 3.00 2.20 2.90 2.00 1.00 2.00 3.20 2.00 2.00 1.80 2.80 2.00 3.00 1.90 2.40 3.00 1.00 2.20 3.30 3.00 2.00 2.30 3.40 3.00 3.00 2.10 2.90 Number of cases read: 9 Number of cases listed: 9
This is called a wide format since the
heights are wide. We may want the data to be long, where each
height is in a separate observation. First, we create a vector of the variable
to be reshaped. Then we create a loop using the index variable and
compute a new variable using the vector, looping as many times as
needed. We save these changes to a new data file and end the loop.
Finally, we get the new data file and list it to make sure that
all went as planned. Note that we use xsave here instead of save.
This is because xsave is not executed until data are read for the next procedure. Hence, it reduces
processing time by consolidating two data passes into one.
vector Aht = ht1 to ht2. loop age = 1 to 2. compute ht = Aht(age). xsave outfile 'c:longex1.sav' /drop ht1 ht2 wt1 wt2. end loop. execute. get file 'c:longex1.sav'. list.FAMID BIRTH AGE HT 1.00 1.00 1.00 2.80 1.00 1.00 2.00 3.40 1.00 2.00 1.00 2.90 1.00 2.00 2.00 3.80 1.00 3.00 1.00 2.20 1.00 3.00 2.00 2.90 2.00 1.00 1.00 2.00 2.00 1.00 2.00 3.20 2.00 2.00 1.00 1.80 2.00 2.00 2.00 2.80 2.00 3.00 1.00 1.90 2.00 3.00 2.00 2.40 3.00 1.00 1.00 2.20 3.00 1.00 2.00 3.30 3.00 2.00 1.00 2.30 3.00 2.00 2.00 3.40 3.00 3.00 1.00 2.10 3.00 3.00 2.00 2.90 Number of cases read: 18 Number of cases listed: 18
Example #2: Two variables
Let’s use the same data file, but with all
of the variables. In this example, we show how to reshape two variables
at a time. Note that you can reshape as many variables as you need by
adding a vector and a compute command for each variable to be
reshaped. You may also want to add those variables to the /drop
subcommand in the aggregate command.
get file 'c:kidshtwt.sav'. list.FAMID BIRTH HT1 HT2 WT1 WT2 1.00 1.00 2.80 3.40 19 28 1.00 2.00 2.90 3.80 21 28 1.00 3.00 2.20 2.90 20 23 2.00 1.00 2.00 3.20 25 30 2.00 2.00 1.80 2.80 20 33 2.00 3.00 1.90 2.40 22 33 3.00 1.00 2.20 3.30 22 28 3.00 2.00 2.30 3.40 20 30 3.00 3.00 2.10 2.90 22 31 Number of cases read: 9 Number of cases listed: 9vector Aht = ht1 to ht2. vector Awt = wt1 to wt2. loop age = 1 to 2. compute ht = Aht(age). compute wt = Awt(age). xsave outfile 'c:longex2.sav' /drop ht1 ht2 wt1 wt2. end loop. execute. get file 'c:longex2.sav'. list.FAMID BIRTH AGE HT WT 1.00 1.00 1.00 2.80 19.00 1.00 1.00 2.00 3.40 28.00 1.00 2.00 1.00 2.90 21.00 1.00 2.00 2.00 3.80 28.00 1.00 3.00 1.00 2.20 20.00 1.00 3.00 2.00 2.90 23.00 2.00 1.00 1.00 2.00 25.00 2.00 1.00 2.00 3.20 30.00 2.00 2.00 1.00 1.80 20.00 2.00 2.00 2.00 2.80 33.00 2.00 3.00 1.00 1.90 22.00 2.00 3.00 2.00 2.40 33.00 3.00 1.00 1.00 2.20 22.00 3.00 1.00 2.00 3.30 28.00 3.00 2.00 1.00 2.30 20.00 3.00 2.00 2.00 3.40 30.00 3.00 3.00 1.00 2.10 22.00 3.00 3.00 2.00 2.90 31.00 Number of cases read: 18 Number of cases listed: 18
Example #3: Modifying numeric suffixes
This example is like the first example in that we are
reshaping only one variable. However, in this example we don’t really
want the loop to start from one as we have done before, but rather start at 96
and go to 98. An extra step is necessary to accomplish this, because in
SPSS the loop has to start with one. Therefore, we add 95 to each value
of year.
get file 'c:faminc.sav'. list.FAMID FAMINC96 FAMINC97 FAMINC98 3.00 75000.00 76000.00 77000.00 1.00 40000.00 40500.00 41000.00 2.00 45000.00 45400.00 45800.00 Number of cases read: 3 Number of cases listed: 3vector Ainc=faminc96 to faminc98. loop i = 1 to 3. compute income=Ainc(i). compute year = 95+i. xsave outfile 'c:widefaminc' /keep famid year income. end loop. execute. get file 'c:widefaminc'. list.FAMID YEAR INCOME 3.00 96.00 75000.00 3.00 97.00 76000.00 3.00 98.00 77000.00 1.00 96.00 40000.00 1.00 97.00 40500.00 1.00 98.00 41000.00 2.00 96.00 45000.00 2.00 97.00 45400.00 2.00 98.00 45800.00 Number of cases read: 9 Number of cases listed: 9
Example #4: String variables and character suffixes
It also is possible to reshape a wide data file to be long when there are character suffixes. Look at the
dmorder file below. Note that we want our long data set to contain a
new string variable called name. To create and/or modify a numeric variable, you
could use the compute command. However, you can only MODIFY a string variable with the
compute command. To CREATE a string variable, you need to use the string command. The syntax for
this command is straight forward: STRING varname (A_), where the _ is the length of
the variable. An example of the use of this command is presented below.
get file 'c:dmorder.sav'. list.FAMID NAMED NAMEM INCD INCM 1.00 Bill Bess 30000.00 15000.00 2.00 Art Amy 22000.00 18000.00 3.00 Paul Pat 25000.00 50000.00 Number of cases read: 3 Number of cases listed: 3vector Aname = named to namem. vector Ainc = incd to incm. string name (A4). loop dadmom = 1 to 2. compute name = Aname(dadmom). compute inc = Ainc(dadmom). xsave outfile 'c:dm1.sav' /keep famid dadmom name inc. end loop. execute. get file 'c:dm1.sav'. list.FAMID DADMOM NAME INC 1.00 1.00 Bill 30000.00 1.00 2.00 Bess 15000.00 2.00 1.00 Art 22000.00 2.00 2.00 Amy 18000.00 3.00 1.00 Paul 25000.00 3.00 2.00 Pat 50000.00 Number of cases read: 6 Number of cases listed: 6
Example #5: Non-contiguous variables
SPSS assumes that the variables to be
reshaped are contiguous (side-by-side) in your data file. If they are not, you will get an error
message when SPSS encounters the vector command. To address this problem, use a
save command with the /keep subcommand, listing the variables in the correct order. The data will be saved with the variables in the order that
they were listed in the /keep subcommand. Notice in the file shown
below that the variables named and namem are not listed next to
each other, nor are incd and incm. Also note that if there
were variables with numeric suffixes, they would have to be contiguous too.
Finally, the variables have to be listed on the vector command in the
order that they appear in the data file. This applies to both numeric and
string variables.
get file 'c:dadmomw.sav'. list.FAMID NAMED INCD NAMEM INCM 1.00 Bill 30000.00 Bess 15000.00 2.00 Art 22000.00 Amy 18000.00 3.00 Paul 25000.00 Pat 50000.00 Number of cases read: 3 Number of cases listed: 3save outfile = "c:dmorder.sav" /keep=famid named namem incd incm. execute. get file 'c:dmorder.sav'. list.FAMID NAMED NAMEM INCD INCM 1.00 Bill Bess 30000.00 15000.00 2.00 Art Amy 22000.00 18000.00 3.00 Paul Pat 25000.00 50000.00 Number of cases read: 3 Number of cases listed: 3
Now that the variables are in the correct
order, we can proceed with the reshaping as before.
vector Aname = named to namem. vector Ainc = incd to incm. string name (A4). loop dadmom = 1 to 2. compute name = Aname(dadmom). compute inc = Ainc(dadmom). xsave outfile 'c:dm.sav' /keep famid dadmom name inc. end loop. execute. get file 'c:dm.sav'. list.FAMID DADMOM NAME INC 1.00 1.00 Bill 30000.00 1.00 2.00 Bess 15000.00 2.00 1.00 Art 22000.00 2.00 2.00 Amy 18000.00 3.00 1.00 Paul 25000.00 3.00 2.00 Pat 50000.00 Number of cases read: 6 Number of cases listed: 6