Benchmarks for text analysis: A response to Budge and Pennings
Budge and Pennings (2007) criticize the “Wordscores” method for computerized content analysis on essentially two grounds. The first is that the best test of Wordscores accuracy is whether it can “reproduce the rich time series produced by the MRG/CMP covering a 50 year period” (Budge and Pennings, 2...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2007
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/soss_research/3977 https://ink.library.smu.edu.sg/context/soss_research/article/5235/viewcontent/ElStud2006_ResponseBP_pv.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | Budge and Pennings (2007) criticize the “Wordscores” method for computerized content analysis on essentially two grounds. The first is that the best test of Wordscores accuracy is whether it can “reproduce the rich time series produced by the MRG/CMP covering a 50 year period” (Budge and Pennings, 2007: 5), which Budge and Pennings claim it does not do. The second is that Wordscores time series estimates, as implemented by Budge and Pennings, yield very little variation around mean scores for the entire time series. In this brief response we make three simple points. |
|---|