An alternative model, which is often recommended for count data, is the square root transformation [29] : (4) This produces less heteroscedastic residuals, though the White [30] and Breusch-Pagan [31] heteroscedasticity tests are extremely significant for all models and so I use robust standard errors clustered by journal for all regressions.
← all excerpts
High-ranked social science journal articles can be identified from early citation information.
1
—
—