rss
J Am Med Inform Assoc 17:3-5 doi:10.1197/jamia.M2853
  • Perspectives on informatics

Use of statistical analysis in the biomedical informatics literature

Table 2

Summary of the statistical methods in JAMIA and IJMI articles (2000–2007)

Type of test JAMIA (n=305) IJMI (n=532)
No (%) No (%)
No statistics 55 (18%) 189 (36%)
Descriptive statistics 216 (71%) 328 (62%)
Elementary statistics 129 (42%) 119 (22%)
 χ2 Analysis 56 (18%) 49 (9%)
 t Test 52 (17%) 44 (8%)
 Kaplan–Meier analysis 2 (1%) 0 (0%)
 Wilcoxon rank sum test 24 (8%) 22 (4%)
 Sign test 1 (0%) 1 (0%)
 Fisher exact test 12 (4%) 9 (2%)
 Analysis of variance 37 (12%) 32 (6%)
 Correlation 51 (17%) 41 (8%)
Multivariable statistics 38 (12%) 31 (6%)
 Multiple logistic regression 23 (8%) 18 (3%)
 Multiple linear regression 15 (5%) 10 (2%)
 Principal-component analysis 4 (1%) 5 (1%)
Other regression analyses 9 (3%) 4 (1%)
Data mining/machine learning 28 (9%) 32 (6%)
 Support vector machines 6 (2%) 4 (1%)
 Bayesian network 12 (4%) 7 (1%)
 Neural networks 1 (0%) 9 (2%)
 Decision trees 8 (3%) 9 (2%)
 Clustering 2 (1%) 4 (1%)
 Hidden Markov Monte Carlo 4 (1%) 5 (1%)
Other statistics 73 (24%) 68 (13%)
 Relative risk/risk ratio 7 (2%) 3 (1%)
 Sensitivity/specificity, precision/recall 63 (21%) 62 (12%)
 Fuzzy logic 1 (0%) 2 (0%)
 Latent semantic analysis 2 (1%) 1 (0%)
 Fourier transform/time series 3 (1%) 1 (0%)
  • Classification format adapted from Windish et al.1

This Article

Access policy for JAMIA

All content published in JAMIA is deposited with PubMed Central by the publisher with a 12 month embargo. Authors/funders may pay an Unlocked fee of $2,000 to make the article free on the JAMIA website and PMC immediately on publication.

All content older than 12 months is freely available on this website.

AMIA members can log in with their JAMIA user name (email address) and password or via the AMIA website.