An Analysis of "Publication Bias" in the Travel, Tourism, and Hospitality Research


YÜKSEL A., TUFAN E.

INTERNATIONAL JOURNAL OF APPLIED BEHAVIORAL ECONOMICS, cilt.6, ss.19-37, 2017 (ESCI İndekslerine Giren Dergi) identifier

  • Cilt numarası: 6 Konu: 4
  • Basım Tarihi: 2017
  • Doi Numarası: 10.4018/ijabe.2017100102
  • Dergi Adı: INTERNATIONAL JOURNAL OF APPLIED BEHAVIORAL ECONOMICS
  • Sayfa Sayıları: ss.19-37

Özet

This article examines whether studies with favorable or statistically significant outcomes are more likely to be published than studies with null results. Should such a publication tendency be in the form of favoring significant findings exist, then the integrity of science, suggestions and conclusions becomes controversial. This also includes those particularly drawn from meta-analyses and systematic reviews. Drawing on a sample of research articles, an examination was undertaken to determine whether studies reporting significant findings were published more. Additional analyses were conducted to examine the validity of reject/support decisions in relation to null hypotheses tested in these studies. The share of the published articles, in which null hypotheses were rejected, was found to be much larger (81%). Interestingly however, calculated power levels and actual samples sizes of these studies were too small to confidently reject/support null hypotheses. Implications for research are discussed in the concluding section of the article.