122:
22:
63:
331:
A jackknife method was applied to meta-analytic data some years later but it does not appear that specialized software was developed for the task. Other commentators have also called for related methods, apparently unaware of the original work. More recent work by a software porting team at Brown
327:
Gee's original software for performing jackknife and combinatorial meta analysis was based on older meta-analytic macros written in the SAS programming language. It was the basis of one report in the area of arthritis treatment. While this software was shared with colleagues informally, it was not
282:
A further implication of this is that arguments over inclusion or exclusion of studies may be moot when the distribution of all possible results is taken into account. A useful tool developed by Gee (reference to come when published) is the "PPES" plot (standing for "Probability of
Positive Effect
390:
Bellamy, N., Campbell, J. and Gee, T. (2005). Can study selection, variable management and time period influence observed effect sizes in systematic reviews of hyaluronan/hylan products?. In: R. Altman, Poster presentations. 10th World
Congress on Osteoarthritis, Massachusetts, U.S.A., (S71-S71).
311:. Where a clear effect is present, this plot should asymptote to near 1.0 fairly rapidly. With this, it is possible then that, for instance, disputes over the inclusion or exclusion of two or three studies out of a dozen or more may be framed in the context of a plot that shows a clear effect for
348:
garbage by a critic, it does offer a way of examining the extent to which those studies may have changed a result. Similarly, it offers no direct solution to the problem of which method to choose for combination or weighting. What it does offer, as noted above, is triangulation, where agreements
318:
It is also possible through CMA to examine the relationship of covariates with effect sizes. For example, if industry funding is suspected as a source of bias, then the proportion of studies in a given subset that were industry funded can be computed and plotted directly against the effect size
203:) is the study of the behaviour of statistical properties of combinations of studies from a meta-analytic dataset (typically in social science research). In an article that develops the notion of "gravity" in the context of meta-analysis, Travis Gee proposed that the
254:
Where it is computationally feasible to obtain all possible combinations, the resulting distribution of statistics is termed "exact CMA." Where the number of possible combinations is prohibitively large, it is termed "approximate CMA."
400:
Marek
Lukacik, MDa, Ronald L. Thomas, PhDb, Jacob V. Aranda, MD, PhDbA Meta-analysis of the Effects of Oral Zinc in the Treatment of Acute and Persistent Diarrhea, Pediatrics Vol. 121 No. 2 February 1, 2008, pp. 326 -336 (doi:
299:. This can be adapted to a "PMES" plot (standing for "Probability of Minimal Effect Size"), where the proportion of studies exceeding some minimal effect size (e.g., SMD = 0.10) is taken for each value of
262:
of adopting a single method and computing a single result, and allows significant triangulation to occur, by computing different indices for each combination and examining whether they all tell the same story.
279:
method to identify the number of intercepts that may be present in the dataset by looking at which studies are included in the local minima that may be obtained through recombination.
319:
estimate. If average age in the various studies was itself fairly variable, then the mean of these means across studies in a given combination can be obtained, and similarly plotted.
258:
CMA makes it possible to study the relative behaviour of different statistics under combinatorial conditions. This differs from the standard approach in
283:
Size", assuming differences are scaled such that larger in a positive direction is desired). For each subset of combinations, where studies are taken
375:
411:
211:
in that article could be extended to examine all possible combinations of studies (where practical) or random subsets of studies (where the
425:
35:
295:
at a time, the proportion of results that show a positive effect size (either WMD or SMD will work) is taken, and this is plotted against
454:
Olkin I, Dahabreh IJ, Trikalinos TA. GOSH - A graphical display of study heterogeneity. Research
Synthesis Methods. 2012;3(3):214-223.
349:
between methods may be obtained, and disagreements between methods understood across the range of possible combinations of studies.
239:
estimates. It is observed that this is a special case of the more general approach of CMA which computes results for
183:
165:
143:
103:
49:
136:
85:
41:
232:
373:
465:"ALLSUBSETS: Stata module to perform all subsets (Combinatorial) meta-analysis in a set of studies"
130:
81:
328:
published. A later meta-analysis applied the concept in the context of the treatment of diarrhea.
433:
484:
147:
426:"Commentary: Heterogeneity in meta-analysis should be expected and appropriately quantified"
8:
73:
368:
Gee, T. (2005) "Capturing study influence: The concept of 'gravity' in meta-analysis",
464:
204:
379:
478:
344:'s problem of "garbage in, garbage out". However, when a class of studies is
341:
272:
259:
212:
208:
271:
An implication of this is that where multiple random intercepts exist, the
276:
275:
within certain combinations will be minimized. CMA can thus be used as a
412:"Statistics Roundtable: The Trusty Jackknife | ASQ"
215:
of the situation made it computationally infeasible).
332:University has implemented the concept in STATA.
476:
364:
362:
359:
84:. There might be a discussion about this on
50:Learn how and when to remove these messages
184:Learn how and when to remove this message
166:Learn how and when to remove this message
104:Learn how and when to remove this message
129:This article includes a list of general
477:
370:Counselling, Psychotherapy, and Health
115:
56:
15:
13:
322:
315:combination of 7 or more studies.
135:it lacks sufficient corresponding
14:
496:
31:This article has multiple issues.
120:
61:
20:
266:
227:objects (studies) are combined
39:or discuss these issues on the
457:
448:
418:
404:
394:
384:
335:
1:
352:
7:
197:Combinatorial meta-analysis
10:
501:
243:studies taken 1, 2, 3 ...
218:
223:In the original article,
401:10.1542/peds.2007-0921)
150:more precise citations.
430:ije.oxfordjournals.org
307: − 1,
291: − 1,
247: − 1,
233:jackknife estimation
74:confusing or unclear
436:on 10 November 2016
340:CMA does not solve
82:clarify the article
467:. 19 October 2012.
378:2006-08-19 at the
205:jackknife methods
194:
193:
186:
176:
175:
168:
114:
113:
106:
54:
492:
469:
468:
461:
455:
452:
446:
445:
443:
441:
432:. Archived from
422:
416:
415:
408:
402:
398:
392:
388:
382:
366:
235:), resulting in
189:
182:
171:
164:
160:
157:
151:
146:this article by
137:inline citations
124:
123:
116:
109:
102:
98:
95:
89:
65:
64:
57:
46:
24:
23:
16:
500:
499:
495:
494:
493:
491:
490:
489:
475:
474:
473:
472:
463:
462:
458:
453:
449:
439:
437:
424:
423:
419:
410:
409:
405:
399:
395:
391:8-11 Dec, 2005.
389:
385:
380:Wayback Machine
367:
360:
355:
338:
325:
323:Implementations
269:
221:
190:
179:
178:
177:
172:
161:
155:
152:
142:Please help to
141:
125:
121:
110:
99:
93:
90:
79:
66:
62:
25:
21:
12:
11:
5:
498:
488:
487:
471:
470:
456:
447:
417:
403:
393:
383:
372:, 1(1), 52–75
357:
356:
354:
351:
337:
334:
324:
321:
268:
265:
231:-1 at a time (
220:
217:
192:
191:
174:
173:
128:
126:
119:
112:
111:
69:
67:
60:
55:
29:
28:
26:
19:
9:
6:
4:
3:
2:
497:
486:
485:Meta-analysis
483:
482:
480:
466:
460:
451:
435:
431:
427:
421:
413:
407:
397:
387:
381:
377:
374:
371:
365:
363:
358:
350:
347:
343:
342:meta-analysis
333:
329:
320:
316:
314:
310:
306:
302:
298:
294:
290:
286:
280:
278:
274:
273:heterogeneity
264:
261:
260:meta-analysis
256:
252:
250:
246:
242:
238:
234:
230:
226:
216:
214:
213:combinatorics
210:
209:meta-analysis
206:
202:
198:
188:
185:
170:
167:
159:
156:February 2010
149:
145:
139:
138:
132:
127:
118:
117:
108:
105:
97:
87:
86:the talk page
83:
77:
75:
70:This article
68:
59:
58:
53:
51:
44:
43:
38:
37:
32:
27:
18:
17:
459:
450:
438:. Retrieved
434:the original
429:
420:
406:
396:
386:
369:
345:
339:
330:
326:
317:
312:
308:
304:
303:= 1, 2, ...
300:
296:
292:
288:
287:= 1, 2, ...
284:
281:
270:
267:Implications
257:
253:
248:
244:
240:
236:
228:
224:
222:
200:
196:
195:
180:
162:
153:
134:
100:
91:
80:Please help
71:
47:
40:
34:
33:Please help
30:
336:Limitations
277:data mining
251:at a time.
207:applied to
148:introducing
440:17 January
353:References
131:references
94:March 2007
76:to readers
36:improve it
42:talk page
479:Category
376:Archived
219:Concept
144:improve
72:may be
346:deemed
133:, but
442:2022
313:any
201:CMA
481::
428:.
361:^
45:.
444:.
414:.
309:k
305:k
301:j
297:j
293:k
289:k
285:j
249:k
245:k
241:k
237:k
229:k
225:k
199:(
187:)
181:(
169:)
163:(
158:)
154:(
140:.
107:)
101:(
96:)
92:(
88:.
78:.
52:)
48:(
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.