At the end of 2025, I pulled the stats for our journals (LibraryPress@UF) on FLVC’s instance of OJS and noticed there was a discrepancy between the stats I was pulling and the stats FLVC collected. The most notable of the discrepancies I found were for two of our journals: EDIS and the Proceedings of the Florida State Horticultural Society (abbreviated as FSHS):
FLVC’s Stats
- EDIS
- Article abstract page views: 2,198,445
- PDF downloads: 622,924
- FSHS
- Article abstract page views: 211,101
- PDF downloads: 196,941
LP@UF’s Stats
- EDIS
- Article abstract page views: 397,525
- PDF downloads: 228,044
- FSHS
- Article abstract page views: 95,951
- PDF downloads: 164,733
I spoke with one of my colleagues at FLVC, and we both ran reports using the same parameters (date range, columns, filters). On our first try, I was still not able to get the same numbers. My colleague suggested not aggregating the stats by month and country, and this solved my problem. We finally got the same stats!
I’m curious as to why aggregating the stats by month and country affect the numbers. I assumed the aggregations were grouping the stats but not actively filtering the stats. Could someone explain what aggregating the stats does? And does removing the aggregation filters give me the most accurate numbers?
I’m using OJS 3.3.0.19.
