The analysis says scientists get less citations after changing fields

The Pivot Size Depends on the Outcome of a Research Move: A Case Study in Obstacles for Scientific Shifts

The study, published today in Nature, also investigates cases in which an external event — such as a disease outbreak — drives the shift in focus. The findings suggest the potential consequences of the recent US research-funding cuts — targeted at research topics including misinformation and the health of minority groups — that could push scientists to change fields.

We used citation-based and non-citation-based measures. Our citation-based measures normalized outcomes for each work by its field and year. The primary citation measure was an indicator for being in the upper fifth percentile among all articles published in the same year. The field designation was the L1 field of research designation, for which there are 154 fields in the Dimensions database. The indicator for being in the upper sixth percentile of citations received among all patents in the same year and technology area is the Cooperative Patent Classification class-level designation.

Given that journals are mostly topic-specific, this shows the scientific fields that a researcher typically draws from, says Dashun Wang, a computational social scientist at Northwestern University in Evanston, Illinois, and a co-author of the study. I cite journals that I have never cited before in order to indicate that I’m doing something very different than what I usually do.

When someone moves out of a trusted area, they are less likely to be recognized according to a science-policy specialist at the University of Manchester, UK. The huge amount of data that the team evaluated set the study apart, he says.

We use binned scatterplots 87 to show potential relationships between pivot size and outcome variables. In Fig. 2a, papers are ordered by average pivot size along the x axis and binned into 20 evenly sized groups. Each marker has a value within each group. Data is presented in Binned Scatterplots. 2c,e,f and 4e,g–j are extended data figs. One sample t-tests can be used to measure mean differences from baseline rates or when evaluating outcomes between high and low pivot-size vigintiles. For simplicity, we report P < 0.001, but note that with observation counts in the millions, these mean tests tend to reject common means with extremely high t-statistics and extremely low P values.

In Figure 4e, the approach is used to split the papers into related or unrelated articles in the year 2020. Similarly, there is also fig. 4g–j presents binned scatterplots to further split the 2020 papers according to the criteria. In a picture. 4k, we account for multivariate controls. We consider the form’s regression.

mi+gamma _rmt+Beta.

The Regressions to Residualize Pivot Size and Impact net of the Control Variables. II. Results and Differences in Differences

There is a relationship between pivot size and outcome with the help of f(Pivot_sizei); and the error term is i. Control variables include fixed effects for average previous impact, author age, team size, number of new collaborators and an indicator variable for funding. In practice, we ran two regressions to residualize pivot size and impact, net of the controls, following the Frisch–Waugh–Lovell theorem. The scatterplot is presented in Figure 4k.

where Postipt is an indicator for the period after the shock. The indicator for being in the treatment group is absorbed with an individual’s fixed effect and so does not appear separately in the regression. Treat_Postipt is an indicator for being in the treatment group after the shock and provides the reported difference-in-differences estimate in Fig. 3. The implications of the external event for pivot size and the reduced form results for impact are shown in Fig. 3b,c. We also present ‘event study’-style difference-in-differences plots in Fig. 3d,e, showing how the treatment effect evolved before and after the retraction date. Here we replace the Treat_Postipt variable and Postipt variable with a series of relative year indicators and their interactions with treatment status.

$${{\rm{Pivot}}_{\rm{size}}}{{\rm{ipt}}}={\mu }{{\rm{i}}}+{\gamma }{{\rm{t}}}+\beta {{\rm{Treat}}_{\rm{Post}}}{{\rm{ipt}}}+\gamma {{\rm{Post}}}{{\rm{ipt}}}+{\varepsilon }{{\rm{ipt}}}$$

Source: The pivot penalty in research

Supplementary Information: Alternative citation-based measures for the locus and publication year of the very highest-income research published by a given author

As presented in the Supplementary Information, we considered numerous alternative citation-based measures. These include smoother (non-binary) outcomes, in which a paper’s citation count is normalized by the mean citation counts to articles in the same field and publication year. We further considered the outcome as the percentile rank of the paper’s citations among all articles published in the same field and year. To examine time frames, we further considered citation counts over two, five and ten-year forward citation windows. Finally, we considered alternative binary indicators to further emphasize the locus of the very highest-income work, defining a ‘hit’ paper as being alternatively in the upper 10, 5, 1, 0.5 or 0.1% percent of all publications in a given field and year. Further details and associated robustness tests for the alternatives are provided in Supplementary section 2.3.2.

If the focal paper drew on journals the same as the author’s previous work, the measure took 0 and 1 respectively. The measure featured in the main text calculates pivoting in the focal paper compared with the past three years of the author’s work. We also calculated our measure by using all previous work of a given author, arriving at similar conclusions (Supplementary section 2.2.1 and Extended Data Fig. 10). Finally, we considered the pivot measure based on the fields of the cited references, rather than their journals, and again found confirmatory results for our main analyses using this alternative measure (Supplementary section 2.2.1 and Supplementary Fig. 3).

varPhi _rmirj