The name is absent



Stata Technical Bulletin


STB-57


will produce


X

myseq

my time

44

10

29decl948

21

12

31decl948

15

13

01janl949

77

23

lljanl949


Saved results

tsmktim saves the items returned by tsset in r().

sbe19.2 Update of tests for publication bias in meta-analysis

Thomas J. Steichen, RJRT, [email protected]

Abstract: Enhancements, changes, and a new option for metabias are described.

Keywords: publication bias, meta-analysis.

This insert documents enhancements and changes to metabias and provides the syntax needed to use a new feature. A full
description of the method and of the operation of the original command and options are given in Steichen (1998). A few revisions
were documented later in Steichen et al. 1998. This updated program does not change the implementation of the underlying
statistical methodology or modify the original operating characteristics of the program; rather, it follows the syntax changes of
Stata version 6.0.

New option

gweight requests that the graphic symbols representing the data in the plot be sized proportional to the inverse variance.

Description

The primary purpose of this version is to revise metabias to meet and to exploit syntax changes in Stata version 6.0. In
addition, some minor deficiencies in the previous implementation have been corrected.

First, metabias previously failed to allow a stratified Egger analysis to report a result for the remaining usable strata when
one or more strata had undefined slopes, instead it reported a missing overall estimate. Such a limitation is not necessary and
has been fixed.

Second, option gweight has been added to allow the graphic symbols to be sized according to the data point’s inverse
variance weight in the optional funnel plot.

Finally, saved results are additionally returned in r(). They are documented below.

Saved Results

The system S_# macros are unchanged. In addition, the saved results are returned in r(). Specifically, metabias saves:

S-I

r(k)

number of studies

S-2

r (score)

Begg’s score

S-3

r(score,sd)

standard deviation of Begg’s score

S-4

r (Begg-p )

Begg’s p value

S _5

r(Begg-pcc)

Begg’s p, continuity corrected

S-6

r (Egger _bc)

Egger’s bias coefficient

S-7

r (Egger _p)

Egger’s p value

S _8

r(effect)

overall effect (log scale)

References

Steichen, T. J. 1998. sbe19: Tests for publication bias in meta-analysis. Stata Technical Bulletin 41: 9-15. Reprinted in Stata Technical Bulletin
Reprints
vol. 7, pp. 125-133.

Steichen, T. J., M. Egger, and J. Sterne. 1998. sbe19.1: Tests for publication bias in meta-analysis. Stata Technical Bulletin 44: 3-4. Reprinted in
Stata Technical Bulletin Reprints vol. 8, pp. 84-85.



More intriguing information

1. WP 92 - An overview of women's work and employment in Azerbaijan
2. Cultural Neuroeconomics of Intertemporal Choice
3. THE MEXICAN HOG INDUSTRY: MOVING BEYOND 2003
4. The name is absent
5. Benchmarking Regional Innovation: A Comparison of Bavaria, Northern Ireland and the Republic of Ireland
6. The name is absent
7. The name is absent
8. Mergers and the changing landscape of commercial banking (Part II)
9. The name is absent
10. Group cooperation, inclusion and disaffected pupils: some responses to informal learning in the music classroom