Example 3

In the third example, we match data from the person file to the author file, and then aggregate information from the manuscript file to determine the review outcome for each version.

1. Are the acceptance rates for initial versions of manuscripts different for men and women?
2. Do acceptance rates vary with the age of the author?

SPSS syntax statement is provided below for each of the three examples.  Non-SPSS users should look at the logic, which can then be replicated with the appropriate SAS of STATA commands.

GET

FILE='Location and name of manuscript file'. 

frequencies variables = RevisionNumber.

/* sort to match in person information: gender and race

/* note use of /Table Command so that person information can be matched for each record where an individual appears as an author'

SORT CASES BY PersonID(A).

MATCH FILES /FILE= *

  /TABLE= 'Location and name of manuscript file'. 

  /BY personID .

EXECUTE.

FREQUENCIES VARIABLES=gender race

  /ORDER=ANALYSIS.

/*  sort so as to be able to match to manuscript to see outcome

sort cases by  ManuscriptNumber RevisionNumber .

SAVE OUTFILE= 'Location and name of manuscript file'. 

  /COMPRESSED.

GET

FILE='Location and name of manuscript file'.   

MATCH FILES /TABLE= *

  /FILE='Location and name of manuscript file'. 

  /BY ManuscriptNumber RevisionNumber .

EXECUTE.

SAVE OUTFILE= 'Location and name of manuscript file'. 

  /COMPRESSED.

/* sample analysis questions

frequencies variables = birthyear.

if ( BirthYear lt 1954)  age = 1.

if ( BirthYear ge 1954 and BirthYear le 1968) age = 2.

if (birthyear gt 1968 ) age = 3.

value labels age 1 'born before 1954'  2 ' born 1954 - 1968'  3 'born after 1968' .

frequencies variables = age.

CROSSTABS

  /TABLES=decision by age

  /FORMAT=AVALUE TABLES

  /STATISTICS=CHISQ

  /CELLS=COUNT COLUMN

  /COUNT ROUND CELL.

if (gender eq 'male'  ) genderN = 1 .

if (gender eq 'female' ) genderN = 2 .

frequencies variables = genderN.

value labels genderN 1 'Male'   2 'Female' .

CROSSTABS

  /TABLES=decision by genderN

  /FORMAT=AVALUE TABLES

  /STATISTICS=CHISQ

  /CELLS=COUNT COLUMN

  /COUNT ROUND CELL.

Using the syntax from this example (Table 7), we see that submissions from female authors are less likely to be accepted and more likely to be rejected than male authors. The differences are relatively small—submissions from male authors were rejected 58.5% of the time as compared to 61.7% for female—however the difference is statistically significant.

Table 7

Using the syntax from this example (Table 8), we see that submissions from younger authors are less likely to be accepted and more likely to be rejected or rejected without review than older authors.

Table 8