SlideShare a Scribd company logo
1 of 61
Master Seminar-II
Meta-Analysis in Agriculture
Aman Vasisht
PGS20AGR8404
Dept. of Agricultural Statistics
UNIVERSITY OF AGRICULTURAL SCIENCES, DHARWAD
COLLEGE OF AGRICULTURE, DHARWAD
OUTLINE
META-ANALYSIS
FORMAT & ASSUMPTIONS OF META-ANALYSIS
CONCEPTS OF META-ANALYSIS
CASE STUDIES
CONCLUSION
REFERENCES
What is Meta-Analysis?
• Analysis of Analyses
• An objective and quantitative methodology for synthesizing
previous studies and research on a particular topic into an overall
finding.
• Meta-Analysis allows us to collect, code, compare or combine
results from different studies and interpret using statistical
methods facilitating statistically guided decisions about the
strength of observed effects and the reliability of results across a
range of studies.
Why Meta-Analysis?
• To provide a summary effect of all previous studies.
The appeal of Meta analysis is that, it leads to effect combines all
the research on one topic into one large study with many studies.
• Overall conclusion of various studies.
• Strength of relationship between studies.
• Relative impact of independent variables.
• Tool for theory development.
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Selecting
area of
interest
and
selecting
variables
Systematic
literature
review
Compile
the results
of these
studies in
a format
Converting
all statistics
to effect
sizes
Meta-
analysis
using
software
Reporting
results
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Selecting
area of
interest
and
selecting
variables
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Systematic
literature
review
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Compile
the results
of these
studies in
a format
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Converting
all statistics
to effect
sizes
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Meta-
analysis
using
software
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Reporting
results
STEP
I
STEP
II
STEP
III
STEP
IV
STEP
V
STEP
VI
Selecting
area of
interest
and
selecting
variables
Systematic
literature
review
Compile
the results
of these
studies in
a format
Converting
all statistics
to effect
sizes
Meta-
analysis
using
software
Reporting
results
Sr. No. Study (Year) N r
1 ABC 2011 40 0.5
2 DEF 2012 90 0.6
3 GHI 2013 250 0.4
4 IJK 2014 400 -0.2
5 LMN 2015 60 -0.3
Sample Format
Sr. No. Study (Year) n1 n2 x1 x2 sd1 sd2
1 ABC 2011
2 DEF 2012
3 GHI 2013
4 IJK 2014
5 LMN 2015
Sr. No. Study (Year) N r
1 ABC 2011 40 0.2
2 DEF 2012 90 0.5
3 GHI 2013 25 0.4
4 IJK 2014 400 0.2
5 LMN 2014 60 0.7
6 OPQ 2015 50 0.45
Example:
Huh! Statistics is boring just with theoretical formulas! I know
you can feel this too! Let’s learn Meta-Analysis practically on
Excel.
Output: Forest Plot
Study weights Confidence Interval
Prediction Interval
Summary Effect
Study N r
ABC 1989 40 0.5
DEF 1990 90 0.6
GHI 1991 250 0.4
IJK 1992 400 0.48
LMN 1993 60 0.35
Study N r
ABC 1989 40 0.5
DEF 1990 90 0.6
GHI 1991 250 0.4
IJK 1992 400 -0.2
LMN 1993 60 -0.3
Concepts in Meta-Analysis
Concepts in Meta-Analysis
Summary Effect
1
Fixed Effect / Random Effect
2
Heterogeneity – Identification and Quantification
3
Confidence Intervals vs Prediction Intervals
4
Subgroup Analysis
5
Meta Regression
6
Assumptions
• An effect is precisely defined, i.e. an independent as well as a
dependent variable are defined, and all studies in the Meta - Analysis
are empirical studies of that effect.
• The type of unit or object in which this effect might occur is
specified e.g., persons, countries, teams specific types of
organizational units) and the domain for which the effect will be
meta-analyzed is clearly delimited.
• Effect size measures in these studies are comparable. Specifically,
• Usually the set of studies that are meta-analyzed are not complete
because some studies have not been published. This is a
problematic assumption..
• All studies are methodologically sound, i.e., data have been
collected from a complete probability sample of a defined
population, measurement has been valid and reliable, and the
statistical analysis has been adequate.
• Assuming that the researcher’s aim is to synthesize empirical
results about the effect in a domain, all empirical studies of the
effect in that domain should have been identified.
Assumptions
Types of Effects
Fixed Effect Random Effect
• Here we assume that there is one
true effect size (hence the term
fixed effect) which underlies all
the studies in the analysis, and
that all differences in observed
effects are due to sampling error.
• No Generalization
• Used rarely and not recommended.
• Here we allow that the true effect
could vary from study to study. If
it were possible to perform an
infinite number of studies, the true
effect sizes for these studies would
be distributed about some mean.
• Purpose is Generalization
• Used frequently and recommended.
Measures of Heterogeneity
• Q-statistic (also referred to as Cochrane’s Q)- It is only a measure of
variation around the average and is not yet a measure of heterogeneity.
• I2 - Proportion of observed variance that reflects real differences in effect
size-expressed as a percentage with a range from 0 to 100%.
• τ2 - Estimate of the variance of the true effect sizes-used to assign weights
to the studies assuming Random Effect.
• Tau (τ) - Estimate of the standard deviation of the distribution of true
effect sizes- used for computing the prediction interval.
Too many formulas.. Let’s go to excel to understand
heterogeneity better
Factors affecting Measures of Dispersions
Range of possible
values
Depends on number of
studies
Depends on scale
Q 1 ≤ Q 
p 0 ≤ p ≤ 1 
T2 0 ≤ T2 
T 0 ≤ T 
I2 0% ≤ I2 ≤ 100%
Confidence Interval vs Prediction Interval
• The confidence interval quantifies the accuracy of the mean, i.e. 95% of
cases the mean effect size falls inside the horizontal black lines.
• Prediction interval addresses the actual dispersion of effect sizes, i.e. 95%
of cases the true effect in a new study will fall inside the horizontal green
lines
Confidence Interval vs Prediction Interval
• The confidence interval quantifies the accuracy of the mean, i.e. 95% of
cases the mean effect size falls inside the horizontal black lines.
• Prediction interval addresses the actual dispersion of effect sizes, i.e. 95%
of cases the true effect in a new study will fall inside the horizontal green
lines
Confidence Interval vs Prediction Interval
• The confidence interval quantifies the accuracy of the mean, i.e. 95% of
cases the mean effect size falls inside the horizontal black lines.
• Prediction interval addresses the actual dispersion of effect sizes, i.e. 95%
of cases the true effect in a new study will fall inside the horizontal green
lines
Confidence Interval
Prediction Interval
Sub-Group Analysis (Non-Metric Moderation)
• When we are working with a single set of studies the effect analysis
assumes that all studies share a common effect size
• When we are working with subgroups, it assumes that all studies within a
subgroup share a common effect size.
• It is determined by I2
Sub-Group Analysis (Non-Metric Moderation)
• When we are working with a single set of studies the effect analysis assumes that all
studies share a common effect size
• When we are working with subgroups, it assumes that all studies within a subgroup
share a common effect size.
• It is determined by I2
Study N r
1. ABC 1989 40 0.5
2. DEF 1990 90 0.6
3. GHI 1991 250 0.4
4. IJK 1992 400 -0.2
5. LMN 1993 60 -0.3
Heterogeneity
Q 78.58
p 0.000
I2 94.91%
T2 0.24
T 0.49
Mod
A
A
A
B
B
A
B
Meta Regression (Metric Moderation)
• When we are working with a single set of studies the effect analysis assumes
that all studies share a common effect size.
• When we are working with meta-regression, it assumes that all studies
which have the same values on the covariates share a common effect
size.
Meta Regression (Metric Moderation)
Sr. No. Study r N Moderator
1 ABC 2007 0.5 40 25
2 DEF 2007 0.6 90 32
3 GHI 2009 0.4 25 28
4 IJK 2011 0.2 400 40
5 LMN 2012 -0.4 60 62
6 OPQ 2014 -0.45 50 73
7 RST 2015 -0.3 100 68
Heterogeneity
Q 92.05
p 0.000
I2 93.48%
T2 0.17
T 0.42
Combined effect size 0.09
R2 91.06%
Effect Size
• Discovering Statistics using SPSS by Andy Field (2013)
explains:
 Just because a test is significant does not mean that it is
important.
 When we measure the size of an effect it is known as
effect size.
 An effect size is simply an objective and (usually)
standardized measure of the magnitude of observed
effect.
Effect Size
• An effect size is a Quantitative measure of the magnitude of a
phenomenon.
• For most types of Effect size, a larger absolute value always indicated
stronger effect (Except odds ratio).
• Effect sizes complement statistical hypothesis testing and play an
important role in Power Analysis, Sample Size Planning and in Meta-
Analysis.
Types of Effect Size
1. Correlation
family
Correlation
coefficient
Coefficient of
determination (R2)
Eta-squared (ƞ2) Omega-squared
(ω2)
Cohen’s f2
Cohen’s q
Used in ANOVA
Effect sizes based
on “variance
explained”
2. Difference
family
Cohen’s d Glass ∆ Hedge’s d 𝜓, root-
mean-square
standardized
effect
Other
metrics
Effect sizes based on
differences between
means
Categorical
family
Cohen’s w Odd ratio
Relative
risk
Cohen’s h
Risk
difference
Effect sizes for
associations among
categorical variables
Interpretation of Effect size:
According to Cohen (1992)
• r = 0.10 (small effect): In this case the effect explains 1% of the total variance.
• r = 0.30 (medium effect): The effect accounts for 9% of the total variance.
• r = 0.50 (large effect): The effect accounts for 25% of the variance.
Formula of Effect Size
Statistics Formula
Correlation r
Independent sample t-test
Paired t-test
Statistics: ANOVA Formula
Eta squared (ƞ2)
Omega squared (ω2)
Cohen’s f2
Again the same
thing…With formulas I
get confused. Show me
in excel with an
example
Effect Size
CASE STUDY-1:
Arbuscular mycorrhizal fungi and antioxidant enzymes
in ameliorating drought stress: a meta-analysis
Chandrasekaran M and Paramasivan M (2022)
Data Source
• Study period: Till 2022 year
• Source: Google Scholar and Web of Science.
• Methodology: Sub-group Meta-Analysis
• Sub-groups : SOD: Superoxide dismutase, CAT: Catalase, POD: Peroxidase,
PPO: Polyphenol peroxidase, APX: Ascorbate peroxidase, GR: Glutathione
reductase, H2O2: Hydrogen peroxide
• Sub-sub groups : Mild (60-80%), Moderate (40-60%) and Severe (0-40%)
based on field capacity.
• Effect size: Response ratio:
ln R = ln
𝑋𝑖
𝑋𝑐
Where, Xi : AMF (Arbuscular mycorrhizal fungi) inoculated plants
(Treatment)
Xj : Without AMF inoculation (control)
Studies identified:
700
Considered: 262
Study excluded:
202
Final studies for
Meta-Analysis: 60
with 505
observations
Results:
AMF inoculation increased
the drought stress alleviation
by 17% more than non-
AMF inoculations.
AMF inoculated plants
significantly reduced the
H2O2 by 20%.
No significant effect for GR
and PPO (11% & 13%,
respectively).
SOD enzyme activity increased
significantly by 32% for AMF
compared to non-AMF
• AMF inoc. increased H2O2 by 1% at mild, decreased by 24% & 25% at moderate and severe, respectively.
• It showed a linear relationship between drought stress and CAT activity in AMF inoculated plants.
• Similar increased linear relationship was seen in APX activity but decreasing linear relationship in POD
activity.
Conclusion of Case Study :
• The Meta-Analysis results showed noticeable up-regulation of the
antioxidant enzyme system (SOD, CAT, POD, and APX) in AMF-
inoculated plants.
• Thereby mediating alleviation of oxidative stress effects of drought stress
through the removal of H2O2.
• Therefore, the necessity to apply eco-friendly approaches, such as AMF-
based bio fertilization is of great importance for agriculture and the
environment.
• Based on enzymatic activities, the effects of drought stress alleviation were
varying with the intensity of drought stress, which were found to be much
more important in predicting plant responses to drought stress than other
variables.
CASE STUDY-2:
Realizing the potential of a low-cost technology to
enhance crop yields: evidence from a meta-analysis of
biofertilizers in India
Praveen K V and A Singh (2019)
Data Source
Studies
identified:
16700
Study titles
screened:
16700
Study
excluded:
16464
Study
abstracts
screened:
236
Final
studies for
Meta-
Analysis:
18
• Study period: 2000 to 2019 year.
• Source: Google Scholar and Consortium of E-Resources in Agriculture (CeRA).
• Methodology: Sub-group Meta-Analysis, Meta regression.
• Sub-groups: Nitrogen fixing biofertilizer, Phosphate solubilizing
biofertilizers, Combined, VAM, Others.
• Moderators: Organic Carbon (%), pH, Total N (kg/ha), Total P (kg/ha).
Forest plot for the effect of different biofertilizer categories on crop yield
Meta Regression
• The analysis suggests that biofertilizer use improves overall yield by 1.59
tonnes per ha.
• The biofertilizers perform better in soils with low pH values, high organic
carbon and low total nitrogen and phosphorus.
• The government must amend the regulatory system to improve the quality
of biofertilizers, and make it easier for farmers to benefit from government
support schemes.
• The government needs to encourage the sector to raise production and
make biofertilizers an integral part of agriculture practices because the
current production is insufficient, even though it has grown rapidly in
recent years.
Conclusion of Case Study :
CASE STUDY-3:
Comparing the Grain Yields of Direct-Seeded
and Transplanted Rice: A Meta-Analysis
Xu Le et al. (2020)
Data Source
• Study period: 1970 to 2017 year.
• Source: Google Scholar and Web of Science.
• Crops: Comparison of direct seeded rice (DSR) and transplanted rice (TPR)
• Methodology: Sub-group Meta-Analysis
• Sub-groups 1: Properties of soil, weed management, tillage and water
management practices.
• Study collected: 53
• Observations used: 440 paired observations
• Effect size: Response ratio:
ln R = ln
𝑋𝐷𝑆𝑅
𝑋𝑇𝑃𝑅
Where, XDSR is mean yield of DSR and XTPR is mean yield of TPR
Results:
DSR yield was 12% lower
than TPR yield on an
average. CI of 13% to 10%
DSR had 7% more panicle
per m2 compared to TPR
Non-significant results
Sub-groups
* Significant at 0.05, ** Significant at 0.01, *** Significant at 0.001
Conclusion of Case Study :
• The meta-analysis indicated that DSR yield was 12% lower than that of TPR
on an average.
• The lower yield of DSR might result from the reduction in spikelet per panicle
and grain weight.
• Weed management, climatic stress and seeding method were the major
contributing factors for the change in yield of DSR from TPR.
• To obtain a comparable yield with TPR, an integrated package of management
technologies should be applied to deal with the major constraints of DSR.
• These results suggest that DSR is a promising alternative planting system in
the face of looming global water scarcity and labour shortages, but more effort
should be made to implement appropriate management strategies and breeding
innovations.
• A Meta-Analysis uses a statistical approach to combine the results to
increase power (over individual studies), improve estimates of the size of the
effect and/or to resolve uncertainty when reports disagree.
• The most important part of Meta-Analysis is systematic review of literature
as publication bias also plays a major role.
• Now, Meta-Analysis is being considered as one of the solid statistical
methods in agriculture.
• Although in cross-sectional agricultural researches there are difficulties due
to different environmental conditions, yet Meta-Analysis has become an
inevitable method, as it can serve as a summary or base for data and results
that can help determine researcher’s path.
Conclusion:
Meta Analysis in Agriculture by Aman Vasisht

More Related Content

What's hot

Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...
Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...
Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...Aaliya Afroz
 
Coefficient of variation
Coefficient of variationCoefficient of variation
Coefficient of variationNadeem Uddin
 
Standardization of rates by Dr. Basil Tumaini
Standardization of rates by Dr. Basil TumainiStandardization of rates by Dr. Basil Tumaini
Standardization of rates by Dr. Basil TumainiBasil Tumaini
 
Investigation Of An Outbreak
Investigation Of An OutbreakInvestigation Of An Outbreak
Investigation Of An Outbreakmeeqat453
 
Forces changing gene frequency
Forces changing gene frequencyForces changing gene frequency
Forces changing gene frequencySyedShaanz
 
Z-scores: Location of Scores and Standardized Distributions
Z-scores: Location of Scores and Standardized DistributionsZ-scores: Location of Scores and Standardized Distributions
Z-scores: Location of Scores and Standardized Distributionsjasondroesch
 
ANOVA & EXPERIMENTAL DESIGNS
ANOVA & EXPERIMENTAL DESIGNSANOVA & EXPERIMENTAL DESIGNS
ANOVA & EXPERIMENTAL DESIGNSvishwanth555
 
050 sampling theory
050 sampling theory050 sampling theory
050 sampling theoryRaj Teotia
 
Kolmogorov Smirnov
Kolmogorov SmirnovKolmogorov Smirnov
Kolmogorov SmirnovRabin BK
 
Colonization of natural enemies
Colonization of natural enemiesColonization of natural enemies
Colonization of natural enemiesIGKV, Raipur
 
Chapter 3 Confidence Interval
Chapter 3 Confidence IntervalChapter 3 Confidence Interval
Chapter 3 Confidence Intervalghalan
 
IPM of Forest Insect Pests
IPM of Forest Insect PestsIPM of Forest Insect Pests
IPM of Forest Insect PestsSyed Ahmed
 
Probit analysis
Probit analysisProbit analysis
Probit analysisPramod935
 
Multiple linear regression II
Multiple linear regression IIMultiple linear regression II
Multiple linear regression IIJames Neill
 
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...World Agroforestry (ICRAF)
 
Introduction to Hypothesis Testing
Introduction to Hypothesis TestingIntroduction to Hypothesis Testing
Introduction to Hypothesis Testingjasondroesch
 
Common sampling techniques in insect pests
Common sampling techniques in insect pestsCommon sampling techniques in insect pests
Common sampling techniques in insect pestsAliRazaIshaq
 
Multiple comparison problem
Multiple comparison problemMultiple comparison problem
Multiple comparison problemJiri Haviger
 

What's hot (20)

Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...
Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...
Colonization of Natural Enemies, Techniques of Release of Natural Enemies, Re...
 
Coefficient of variation
Coefficient of variationCoefficient of variation
Coefficient of variation
 
Standardization of rates by Dr. Basil Tumaini
Standardization of rates by Dr. Basil TumainiStandardization of rates by Dr. Basil Tumaini
Standardization of rates by Dr. Basil Tumaini
 
Investigation Of An Outbreak
Investigation Of An OutbreakInvestigation Of An Outbreak
Investigation Of An Outbreak
 
Forces changing gene frequency
Forces changing gene frequencyForces changing gene frequency
Forces changing gene frequency
 
Z-scores: Location of Scores and Standardized Distributions
Z-scores: Location of Scores and Standardized DistributionsZ-scores: Location of Scores and Standardized Distributions
Z-scores: Location of Scores and Standardized Distributions
 
ANOVA & EXPERIMENTAL DESIGNS
ANOVA & EXPERIMENTAL DESIGNSANOVA & EXPERIMENTAL DESIGNS
ANOVA & EXPERIMENTAL DESIGNS
 
050 sampling theory
050 sampling theory050 sampling theory
050 sampling theory
 
Kolmogorov Smirnov
Kolmogorov SmirnovKolmogorov Smirnov
Kolmogorov Smirnov
 
Colonization of natural enemies
Colonization of natural enemiesColonization of natural enemies
Colonization of natural enemies
 
Coefficient of Variation
Coefficient of VariationCoefficient of Variation
Coefficient of Variation
 
Chapter 3 Confidence Interval
Chapter 3 Confidence IntervalChapter 3 Confidence Interval
Chapter 3 Confidence Interval
 
IPM of Forest Insect Pests
IPM of Forest Insect PestsIPM of Forest Insect Pests
IPM of Forest Insect Pests
 
Probit analysis
Probit analysisProbit analysis
Probit analysis
 
Multiple linear regression II
Multiple linear regression IIMultiple linear regression II
Multiple linear regression II
 
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...
Genotype to phenotype forest tree genomics: genome sequencing (de novo and re...
 
T test for one sample mean
T test for one sample meanT test for one sample mean
T test for one sample mean
 
Introduction to Hypothesis Testing
Introduction to Hypothesis TestingIntroduction to Hypothesis Testing
Introduction to Hypothesis Testing
 
Common sampling techniques in insect pests
Common sampling techniques in insect pestsCommon sampling techniques in insect pests
Common sampling techniques in insect pests
 
Multiple comparison problem
Multiple comparison problemMultiple comparison problem
Multiple comparison problem
 

Similar to Meta Analysis in Agriculture by Aman Vasisht

Sample determinants and size
Sample determinants and sizeSample determinants and size
Sample determinants and sizeTarek Tawfik Amin
 
Sample-size-comprehensive.pptx
Sample-size-comprehensive.pptxSample-size-comprehensive.pptx
Sample-size-comprehensive.pptxssuser4eb7dd
 
Research and Scientific Journal Publication support services | Research pape...
 Research and Scientific Journal Publication support services | Research pape... Research and Scientific Journal Publication support services | Research pape...
Research and Scientific Journal Publication support services | Research pape...Pubrica
 
2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbordrgveroniki
 
Cochrane Collaboration
Cochrane CollaborationCochrane Collaboration
Cochrane CollaborationNinian Peckitt
 
Sample size calculation
Sample  size calculationSample  size calculation
Sample size calculationSwati Singh
 
Meta-Analysis and Research Synthesis
Meta-Analysis and Research SynthesisMeta-Analysis and Research Synthesis
Meta-Analysis and Research SynthesisTomás de la Rosa
 
RSS 2008 - meta-analyis when assumptions are violated
RSS 2008 - meta-analyis when assumptions are violatedRSS 2008 - meta-analyis when assumptions are violated
RSS 2008 - meta-analyis when assumptions are violatedEvangelos Kontopantelis
 
PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,Naveen K L
 
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.ppt
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.pptIntroduction_klsfnsfsnfsgnkgni _to_meta-analysis.ppt
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.pptAnnaMarieAndalRanill
 
Meta analysis
Meta analysisMeta analysis
Meta analysisJunaidAKG
 
1625941932480.pptx
1625941932480.pptx1625941932480.pptx
1625941932480.pptxMathiQueeny
 
One Sample t test.pptx
One Sample t test.pptxOne Sample t test.pptx
One Sample t test.pptxletbestrong
 
Planning of the experiments in research
Planning of the experiments in researchPlanning of the experiments in research
Planning of the experiments in researchpbbharate
 
Bio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchBio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchShinjan Patra
 
Introduction to Meta-analysis - Dr Moses Ocan
Introduction to Meta-analysis - Dr Moses OcanIntroduction to Meta-analysis - Dr Moses Ocan
Introduction to Meta-analysis - Dr Moses OcanACSRM
 

Similar to Meta Analysis in Agriculture by Aman Vasisht (20)

Sample determinants and size
Sample determinants and sizeSample determinants and size
Sample determinants and size
 
Sample-size-comprehensive.pptx
Sample-size-comprehensive.pptxSample-size-comprehensive.pptx
Sample-size-comprehensive.pptx
 
Research and Scientific Journal Publication support services | Research pape...
 Research and Scientific Journal Publication support services | Research pape... Research and Scientific Journal Publication support services | Research pape...
Research and Scientific Journal Publication support services | Research pape...
 
2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord
 
Cochrane Collaboration
Cochrane CollaborationCochrane Collaboration
Cochrane Collaboration
 
Sample size calculation
Sample  size calculationSample  size calculation
Sample size calculation
 
Meta-Analysis and Research Synthesis
Meta-Analysis and Research SynthesisMeta-Analysis and Research Synthesis
Meta-Analysis and Research Synthesis
 
RSS 2008 - meta-analyis when assumptions are violated
RSS 2008 - meta-analyis when assumptions are violatedRSS 2008 - meta-analyis when assumptions are violated
RSS 2008 - meta-analyis when assumptions are violated
 
PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,
 
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.ppt
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.pptIntroduction_klsfnsfsnfsgnkgni _to_meta-analysis.ppt
Introduction_klsfnsfsnfsgnkgni _to_meta-analysis.ppt
 
Validity andreliability
Validity andreliabilityValidity andreliability
Validity andreliability
 
Meta analysis
Meta analysisMeta analysis
Meta analysis
 
1625941932480.pptx
1625941932480.pptx1625941932480.pptx
1625941932480.pptx
 
One Sample t test.pptx
One Sample t test.pptxOne Sample t test.pptx
One Sample t test.pptx
 
Planning of the experiments in research
Planning of the experiments in researchPlanning of the experiments in research
Planning of the experiments in research
 
Bio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchBio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical research
 
Sample size determination
Sample size determinationSample size determination
Sample size determination
 
Introduction to Meta-analysis - Dr Moses Ocan
Introduction to Meta-analysis - Dr Moses OcanIntroduction to Meta-analysis - Dr Moses Ocan
Introduction to Meta-analysis - Dr Moses Ocan
 
Meta analysis with R
Meta analysis with RMeta analysis with R
Meta analysis with R
 
Sample and effect size
Sample and effect sizeSample and effect size
Sample and effect size
 

Recently uploaded

GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]📊 Markus Baersch
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceSapana Sha
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
How we prevented account sharing with MFA
How we prevented account sharing with MFAHow we prevented account sharing with MFA
How we prevented account sharing with MFAAndrei Kaleshka
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.natarajan8993
 
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degreeyuu sss
 
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样vhwb25kk
 
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...soniya singh
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queensdataanalyticsqueen03
 
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort servicejennyeacort
 
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一F La
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home ServiceSapana Sha
 
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理科罗拉多大学波尔得分校毕业证学位证成绩单-可办理
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理e4aez8ss
 
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝soniya singh
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degreeyuu sss
 
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptdokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptSonatrach
 
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Sapana Sha
 

Recently uploaded (20)

GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts Service
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
How we prevented account sharing with MFA
How we prevented account sharing with MFAHow we prevented account sharing with MFA
How we prevented account sharing with MFA
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.
 
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree
毕业文凭制作#回国入职#diploma#degree澳洲中央昆士兰大学毕业证成绩单pdf电子版制作修改#毕业文凭制作#回国入职#diploma#degree
 
Call Girls in Saket 99530🔝 56974 Escort Service
Call Girls in Saket 99530🔝 56974 Escort ServiceCall Girls in Saket 99530🔝 56974 Escort Service
Call Girls in Saket 99530🔝 56974 Escort Service
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样
1:1定制(UQ毕业证)昆士兰大学毕业证成绩单修改留信学历认证原版一模一样
 
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queens
 
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
 
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service
 
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理科罗拉多大学波尔得分校毕业证学位证成绩单-可办理
科罗拉多大学波尔得分校毕业证学位证成绩单-可办理
 
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
 
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptdokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
 
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
 

Meta Analysis in Agriculture by Aman Vasisht

  • 1.
  • 2. Master Seminar-II Meta-Analysis in Agriculture Aman Vasisht PGS20AGR8404 Dept. of Agricultural Statistics UNIVERSITY OF AGRICULTURAL SCIENCES, DHARWAD COLLEGE OF AGRICULTURE, DHARWAD
  • 3. OUTLINE META-ANALYSIS FORMAT & ASSUMPTIONS OF META-ANALYSIS CONCEPTS OF META-ANALYSIS CASE STUDIES CONCLUSION REFERENCES
  • 4. What is Meta-Analysis? • Analysis of Analyses • An objective and quantitative methodology for synthesizing previous studies and research on a particular topic into an overall finding. • Meta-Analysis allows us to collect, code, compare or combine results from different studies and interpret using statistical methods facilitating statistically guided decisions about the strength of observed effects and the reliability of results across a range of studies.
  • 5. Why Meta-Analysis? • To provide a summary effect of all previous studies. The appeal of Meta analysis is that, it leads to effect combines all the research on one topic into one large study with many studies. • Overall conclusion of various studies. • Strength of relationship between studies. • Relative impact of independent variables. • Tool for theory development.
  • 6. STEP I STEP II STEP III STEP IV STEP V STEP VI Selecting area of interest and selecting variables Systematic literature review Compile the results of these studies in a format Converting all statistics to effect sizes Meta- analysis using software Reporting results
  • 13. STEP I STEP II STEP III STEP IV STEP V STEP VI Selecting area of interest and selecting variables Systematic literature review Compile the results of these studies in a format Converting all statistics to effect sizes Meta- analysis using software Reporting results
  • 14. Sr. No. Study (Year) N r 1 ABC 2011 40 0.5 2 DEF 2012 90 0.6 3 GHI 2013 250 0.4 4 IJK 2014 400 -0.2 5 LMN 2015 60 -0.3 Sample Format Sr. No. Study (Year) n1 n2 x1 x2 sd1 sd2 1 ABC 2011 2 DEF 2012 3 GHI 2013 4 IJK 2014 5 LMN 2015
  • 15. Sr. No. Study (Year) N r 1 ABC 2011 40 0.2 2 DEF 2012 90 0.5 3 GHI 2013 25 0.4 4 IJK 2014 400 0.2 5 LMN 2014 60 0.7 6 OPQ 2015 50 0.45 Example: Huh! Statistics is boring just with theoretical formulas! I know you can feel this too! Let’s learn Meta-Analysis practically on Excel.
  • 16. Output: Forest Plot Study weights Confidence Interval Prediction Interval Summary Effect
  • 17. Study N r ABC 1989 40 0.5 DEF 1990 90 0.6 GHI 1991 250 0.4 IJK 1992 400 0.48 LMN 1993 60 0.35 Study N r ABC 1989 40 0.5 DEF 1990 90 0.6 GHI 1991 250 0.4 IJK 1992 400 -0.2 LMN 1993 60 -0.3
  • 19. Concepts in Meta-Analysis Summary Effect 1 Fixed Effect / Random Effect 2 Heterogeneity – Identification and Quantification 3 Confidence Intervals vs Prediction Intervals 4 Subgroup Analysis 5 Meta Regression 6
  • 20. Assumptions • An effect is precisely defined, i.e. an independent as well as a dependent variable are defined, and all studies in the Meta - Analysis are empirical studies of that effect. • The type of unit or object in which this effect might occur is specified e.g., persons, countries, teams specific types of organizational units) and the domain for which the effect will be meta-analyzed is clearly delimited. • Effect size measures in these studies are comparable. Specifically,
  • 21. • Usually the set of studies that are meta-analyzed are not complete because some studies have not been published. This is a problematic assumption.. • All studies are methodologically sound, i.e., data have been collected from a complete probability sample of a defined population, measurement has been valid and reliable, and the statistical analysis has been adequate. • Assuming that the researcher’s aim is to synthesize empirical results about the effect in a domain, all empirical studies of the effect in that domain should have been identified. Assumptions
  • 22. Types of Effects Fixed Effect Random Effect • Here we assume that there is one true effect size (hence the term fixed effect) which underlies all the studies in the analysis, and that all differences in observed effects are due to sampling error. • No Generalization • Used rarely and not recommended. • Here we allow that the true effect could vary from study to study. If it were possible to perform an infinite number of studies, the true effect sizes for these studies would be distributed about some mean. • Purpose is Generalization • Used frequently and recommended.
  • 23. Measures of Heterogeneity • Q-statistic (also referred to as Cochrane’s Q)- It is only a measure of variation around the average and is not yet a measure of heterogeneity. • I2 - Proportion of observed variance that reflects real differences in effect size-expressed as a percentage with a range from 0 to 100%. • τ2 - Estimate of the variance of the true effect sizes-used to assign weights to the studies assuming Random Effect. • Tau (τ) - Estimate of the standard deviation of the distribution of true effect sizes- used for computing the prediction interval. Too many formulas.. Let’s go to excel to understand heterogeneity better
  • 24. Factors affecting Measures of Dispersions Range of possible values Depends on number of studies Depends on scale Q 1 ≤ Q  p 0 ≤ p ≤ 1  T2 0 ≤ T2  T 0 ≤ T  I2 0% ≤ I2 ≤ 100%
  • 25. Confidence Interval vs Prediction Interval • The confidence interval quantifies the accuracy of the mean, i.e. 95% of cases the mean effect size falls inside the horizontal black lines. • Prediction interval addresses the actual dispersion of effect sizes, i.e. 95% of cases the true effect in a new study will fall inside the horizontal green lines
  • 26. Confidence Interval vs Prediction Interval • The confidence interval quantifies the accuracy of the mean, i.e. 95% of cases the mean effect size falls inside the horizontal black lines. • Prediction interval addresses the actual dispersion of effect sizes, i.e. 95% of cases the true effect in a new study will fall inside the horizontal green lines
  • 27. Confidence Interval vs Prediction Interval • The confidence interval quantifies the accuracy of the mean, i.e. 95% of cases the mean effect size falls inside the horizontal black lines. • Prediction interval addresses the actual dispersion of effect sizes, i.e. 95% of cases the true effect in a new study will fall inside the horizontal green lines Confidence Interval Prediction Interval
  • 28. Sub-Group Analysis (Non-Metric Moderation) • When we are working with a single set of studies the effect analysis assumes that all studies share a common effect size • When we are working with subgroups, it assumes that all studies within a subgroup share a common effect size. • It is determined by I2
  • 29. Sub-Group Analysis (Non-Metric Moderation) • When we are working with a single set of studies the effect analysis assumes that all studies share a common effect size • When we are working with subgroups, it assumes that all studies within a subgroup share a common effect size. • It is determined by I2 Study N r 1. ABC 1989 40 0.5 2. DEF 1990 90 0.6 3. GHI 1991 250 0.4 4. IJK 1992 400 -0.2 5. LMN 1993 60 -0.3 Heterogeneity Q 78.58 p 0.000 I2 94.91% T2 0.24 T 0.49 Mod A A A B B
  • 30. A B
  • 31. Meta Regression (Metric Moderation) • When we are working with a single set of studies the effect analysis assumes that all studies share a common effect size. • When we are working with meta-regression, it assumes that all studies which have the same values on the covariates share a common effect size.
  • 32. Meta Regression (Metric Moderation) Sr. No. Study r N Moderator 1 ABC 2007 0.5 40 25 2 DEF 2007 0.6 90 32 3 GHI 2009 0.4 25 28 4 IJK 2011 0.2 400 40 5 LMN 2012 -0.4 60 62 6 OPQ 2014 -0.45 50 73 7 RST 2015 -0.3 100 68 Heterogeneity Q 92.05 p 0.000 I2 93.48% T2 0.17 T 0.42 Combined effect size 0.09 R2 91.06%
  • 33. Effect Size • Discovering Statistics using SPSS by Andy Field (2013) explains:  Just because a test is significant does not mean that it is important.  When we measure the size of an effect it is known as effect size.  An effect size is simply an objective and (usually) standardized measure of the magnitude of observed effect.
  • 34. Effect Size • An effect size is a Quantitative measure of the magnitude of a phenomenon. • For most types of Effect size, a larger absolute value always indicated stronger effect (Except odds ratio). • Effect sizes complement statistical hypothesis testing and play an important role in Power Analysis, Sample Size Planning and in Meta- Analysis.
  • 35. Types of Effect Size 1. Correlation family Correlation coefficient Coefficient of determination (R2) Eta-squared (ƞ2) Omega-squared (ω2) Cohen’s f2 Cohen’s q Used in ANOVA Effect sizes based on “variance explained”
  • 36. 2. Difference family Cohen’s d Glass ∆ Hedge’s d 𝜓, root- mean-square standardized effect Other metrics Effect sizes based on differences between means
  • 37. Categorical family Cohen’s w Odd ratio Relative risk Cohen’s h Risk difference Effect sizes for associations among categorical variables Interpretation of Effect size: According to Cohen (1992) • r = 0.10 (small effect): In this case the effect explains 1% of the total variance. • r = 0.30 (medium effect): The effect accounts for 9% of the total variance. • r = 0.50 (large effect): The effect accounts for 25% of the variance.
  • 38. Formula of Effect Size Statistics Formula Correlation r Independent sample t-test Paired t-test Statistics: ANOVA Formula Eta squared (ƞ2) Omega squared (ω2) Cohen’s f2 Again the same thing…With formulas I get confused. Show me in excel with an example Effect Size
  • 39. CASE STUDY-1: Arbuscular mycorrhizal fungi and antioxidant enzymes in ameliorating drought stress: a meta-analysis Chandrasekaran M and Paramasivan M (2022)
  • 40. Data Source • Study period: Till 2022 year • Source: Google Scholar and Web of Science. • Methodology: Sub-group Meta-Analysis • Sub-groups : SOD: Superoxide dismutase, CAT: Catalase, POD: Peroxidase, PPO: Polyphenol peroxidase, APX: Ascorbate peroxidase, GR: Glutathione reductase, H2O2: Hydrogen peroxide • Sub-sub groups : Mild (60-80%), Moderate (40-60%) and Severe (0-40%) based on field capacity. • Effect size: Response ratio: ln R = ln 𝑋𝑖 𝑋𝑐 Where, Xi : AMF (Arbuscular mycorrhizal fungi) inoculated plants (Treatment) Xj : Without AMF inoculation (control)
  • 41. Studies identified: 700 Considered: 262 Study excluded: 202 Final studies for Meta-Analysis: 60 with 505 observations Results: AMF inoculation increased the drought stress alleviation by 17% more than non- AMF inoculations. AMF inoculated plants significantly reduced the H2O2 by 20%. No significant effect for GR and PPO (11% & 13%, respectively). SOD enzyme activity increased significantly by 32% for AMF compared to non-AMF
  • 42. • AMF inoc. increased H2O2 by 1% at mild, decreased by 24% & 25% at moderate and severe, respectively. • It showed a linear relationship between drought stress and CAT activity in AMF inoculated plants. • Similar increased linear relationship was seen in APX activity but decreasing linear relationship in POD activity.
  • 43. Conclusion of Case Study : • The Meta-Analysis results showed noticeable up-regulation of the antioxidant enzyme system (SOD, CAT, POD, and APX) in AMF- inoculated plants. • Thereby mediating alleviation of oxidative stress effects of drought stress through the removal of H2O2. • Therefore, the necessity to apply eco-friendly approaches, such as AMF- based bio fertilization is of great importance for agriculture and the environment. • Based on enzymatic activities, the effects of drought stress alleviation were varying with the intensity of drought stress, which were found to be much more important in predicting plant responses to drought stress than other variables.
  • 44. CASE STUDY-2: Realizing the potential of a low-cost technology to enhance crop yields: evidence from a meta-analysis of biofertilizers in India Praveen K V and A Singh (2019)
  • 45. Data Source Studies identified: 16700 Study titles screened: 16700 Study excluded: 16464 Study abstracts screened: 236 Final studies for Meta- Analysis: 18 • Study period: 2000 to 2019 year. • Source: Google Scholar and Consortium of E-Resources in Agriculture (CeRA). • Methodology: Sub-group Meta-Analysis, Meta regression. • Sub-groups: Nitrogen fixing biofertilizer, Phosphate solubilizing biofertilizers, Combined, VAM, Others. • Moderators: Organic Carbon (%), pH, Total N (kg/ha), Total P (kg/ha).
  • 46. Forest plot for the effect of different biofertilizer categories on crop yield
  • 47.
  • 48.
  • 49.
  • 50.
  • 52.
  • 53. • The analysis suggests that biofertilizer use improves overall yield by 1.59 tonnes per ha. • The biofertilizers perform better in soils with low pH values, high organic carbon and low total nitrogen and phosphorus. • The government must amend the regulatory system to improve the quality of biofertilizers, and make it easier for farmers to benefit from government support schemes. • The government needs to encourage the sector to raise production and make biofertilizers an integral part of agriculture practices because the current production is insufficient, even though it has grown rapidly in recent years. Conclusion of Case Study :
  • 54. CASE STUDY-3: Comparing the Grain Yields of Direct-Seeded and Transplanted Rice: A Meta-Analysis Xu Le et al. (2020)
  • 55. Data Source • Study period: 1970 to 2017 year. • Source: Google Scholar and Web of Science. • Crops: Comparison of direct seeded rice (DSR) and transplanted rice (TPR) • Methodology: Sub-group Meta-Analysis • Sub-groups 1: Properties of soil, weed management, tillage and water management practices. • Study collected: 53 • Observations used: 440 paired observations • Effect size: Response ratio: ln R = ln 𝑋𝐷𝑆𝑅 𝑋𝑇𝑃𝑅 Where, XDSR is mean yield of DSR and XTPR is mean yield of TPR
  • 56. Results: DSR yield was 12% lower than TPR yield on an average. CI of 13% to 10% DSR had 7% more panicle per m2 compared to TPR Non-significant results
  • 58. * Significant at 0.05, ** Significant at 0.01, *** Significant at 0.001
  • 59. Conclusion of Case Study : • The meta-analysis indicated that DSR yield was 12% lower than that of TPR on an average. • The lower yield of DSR might result from the reduction in spikelet per panicle and grain weight. • Weed management, climatic stress and seeding method were the major contributing factors for the change in yield of DSR from TPR. • To obtain a comparable yield with TPR, an integrated package of management technologies should be applied to deal with the major constraints of DSR. • These results suggest that DSR is a promising alternative planting system in the face of looming global water scarcity and labour shortages, but more effort should be made to implement appropriate management strategies and breeding innovations.
  • 60. • A Meta-Analysis uses a statistical approach to combine the results to increase power (over individual studies), improve estimates of the size of the effect and/or to resolve uncertainty when reports disagree. • The most important part of Meta-Analysis is systematic review of literature as publication bias also plays a major role. • Now, Meta-Analysis is being considered as one of the solid statistical methods in agriculture. • Although in cross-sectional agricultural researches there are difficulties due to different environmental conditions, yet Meta-Analysis has become an inevitable method, as it can serve as a summary or base for data and results that can help determine researcher’s path. Conclusion: