Example 1

In the first example we aggregate all versions of manuscripts at the manuscript level to track trajectories of particular types of manuscripts. Information for each manuscript is then added to the record for each author.  Finally author characteristics are added in from the person file.

1. What type of topics are most likely to be accepted at ASR? Use titles to develop topic codes. How has this changed over time?
2. What type of authors are most successful?
3. What is the average submission to final decision process for manuscripts look like?

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

Unit of Analysis:

Manuscript submissions with one record for each author, with information on each round of review.  Looking at those cases where the original submission happened in 1990 or later.  Some manuscripts that were submitted before 1990 were still in review during 1990.  These cases are omitted as the original submission is left censored.

* Begin with manuscript file and specify local directory

GET

FILE='Location and name of manuscript file'.   

SORT CASES BY ManuscriptNumber(A) RevisionNumber(A).

/* determine maximum number of revisions 7 including R(0)

FREQUENCIES VARIABLES=RevisionNumber  /ORDER=ANALYSIS.

/* Create and populate variables  for aggregation one set for decisions, one for title

string decision0 decision1 decision2 decision3 decision4 decision5 decision6 (A17).

if ( RevisionNumber eq '(R0)') decision0 = Decision.

if ( RevisionNumber eq '(R1)') decision1 = Decision.

if ( RevisionNumber eq '(R2)') decision2 = Decision.

if ( RevisionNumber eq '(R3)') decision3 = Decision.

if ( RevisionNumber eq '(R4)') decision4 = Decision.

if ( RevisionNumber eq '(R5)') decision5 = Decision.

if ( RevisionNumber eq '(R6)') decision6 = Decision.

string title0 title1 title2 title3 title4 title5 title6 (A17).

if ( RevisionNumber eq '(R0)') title0 = title.

if ( RevisionNumber eq '(R1)') title1 = title.

if ( RevisionNumber eq '(R2)') title2 = title.

if ( RevisionNumber eq '(R3)') title3 = title.

if ( RevisionNumber eq '(R4)') title4 = title.

if ( RevisionNumber eq '(R5)') title5 = title.

if ( RevisionNumber eq '(R6)') title6 = title.

Execute.

/* aggregate and save manuscript file with 1 record for all versions

DATASET DECLARE Manuscript_aggregate.

AGGREGATE

  /OUTFILE='Manuscript_aggregate'

  /BREAK=ManuscriptNumber

  /decision0=MAX(decision0)

  /decision1=MAX(decision1)

  /decision2=MAX(decision2)

  /decision3=MAX(decision3)

  /decision4=MAX(decision4)

  /decision5=MAX(decision5)

  /decision6=MAX(decision6)

  /title0=MAX(title0)

  /title1=MAX(title1)

  /title2=MAX(title2)

  /title3=MAX(title3)

  /title4=MAX(title4)

  /title5=MAX(title5)

  /title6=MAX(title6)

  /ReceivedDate=FIRST(ReceivedDate)

  /DecisionDate=LAST(DecisionDate).

Execute.

SAVE

FILE='Location and name of manuscript file'. 

/COMPRESSED.

/* now get author information so person number can be linked to manuscripts

GET

FILE='Location and name of manuscript file'. 

/* match author to manuscript aggregated version

SORT CASES BY ManuscriptNumber(A) PersonID RevisionNumber(A).

MATCH FILES /FILE= *

  /TABLE= 'Location and name of manuscript file'

  /BY ManuscriptNumber .

EXECUTE.

frequencies variables =  RevisionNumber .

/* now select the original submissions

select if ( RevisionNumber eq '(R0)') .

frequencies variables =  RevisionNumber .

SORT CASES BY PersonID .

MATCH FILES /FILE= *

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

  /BY personID .

EXECUTE.

compute nversions = 0.

do repeat decision = decision0 decision1 decision2 decision3 decision4 decision5 decision6.

if (length(decision) gt 0) nversions = nversions + 1.

end repeat.

frequencies variables = nversions.

string decision lastdecision (A17).

Execute.

if (nversions eq 1) lastdecision = decision0.

if (nversions eq 2) lastdecision = decision1.

if (nversions eq 3) lastdecision = decision2.

if (nversions eq 4) lastdecision = decision3.

if (nversions eq 5) lastdecision = decision4.

if (nversions eq 6) lastdecision = decision5.

if (nversions eq 7) lastdecision = decision6.

/* sample analysis questions

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.

if (gender eq 'male'  ) genderN = 1 .

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

frequencies variables = genderN.

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

frequencies variables = lastdecision.

* crosstab check should have 14,001 cases

CROSSTABS

  /TABLES=lastdecision by genderN

  /FORMAT=AVALUE TABLES

  /STATISTICS=CHISQ

  /CELLS=COUNT COLUMN

  /COUNT ROUND CELL.

CROSSTABS

  /TABLES=lastdecision by age

  /FORMAT=AVALUE TABLES

  /STATISTICS=CHISQ

  /CELLS=COUNT COLUMN

  /COUNT ROUND CELL.

Table 3a shows that only 13.7% of manuscripts submitted to ASR during this time period were accepted.  This number could go slightly higher as the final decision on 2.1% of the manuscripts is a Revise and Resubmit.

Table 3a

Table 3a shows that original submissions from male authors were significantly more likely to be ultimately accepted than those submitted by female authors.

Table 3b

Manuscripts from older authors were significantly more likely to ultimately be accepted than those from younger authors—this difference is not only statistically significant but also relatively large. (Table 3c)

Table 3c