Impact Factors in Prediction of Future Citations

Although many of us have heard mention of the use of the journal impact factor as an indicator of the impact or quality of scholars' work, little empirical work has been done on this application to date. Does it make sense for academic employment decision makers (e.g. promotion and tenure committees, deans, etc.) to use the impact factor score as an proxy indicator of the eventual impact of a scholar’s work in those instances where that work is so new that the pattern of citations it will attract is unknown?

Holden et al. 2006 explored this question using a random sample of 323 articles selected from 17 social work journals listed in Journal Citation Reports. In this sample from the 1992-1994 period, they found that the impact factor of the journal the article was published in predicted less than 12% of the variance in outcome (r=0.4) at either four or ten years in the future. Given these short and long term results, the authors concluded that the practice of using the impact factor as a proxy indicator of article impact is not supported at this time and further research may be needed.

However, relevance to epidemiology journals is unknown. Furthermore, the modest correlation of current journal impact factor in prediction future citation may be due to temporal trends in the prominence of a journal itself.

Overall, journal impact factors have a list of limitations to overcome.


Holden, G., Rosenberg, G., Barker, K. & Onghena, P. (2006). An assessment of the predictive validity of impact factor scores: Implications for academic employment decisions in social work. Research on Social Work Practice, 16, 6, 613-624.