Table 3

Simplified table representing the identification key of possible analytical methods (Gomez and Gomez, 1984; Siegel and Castellan Jr, 1988; Cady, 1991; Federer, 1999; De’ath, 2002; Baayen et al., 2008; Zuur et al., 200; 9; Dagnelie, 2012; Payne, 2015) based on three generic experimental questions (column 1). This table does not enable one to answer agronomic questions on a case-by-case basis, but it does provide guidance and access to the main families of data analysis whenever possible. It is read from left to right. The user chooses at each step the line that corresponds to a given experimental situation. The proposed analyses can be undertaken only after having verified that the data meet their validity conditions. A good use of this table begins with the identification of the experiment’s statistical terms: factors, modalities, measurement scale, blocks, control.

Objective Key 1 Key 2 Main families
Compare groups or systems 1 factor 2 groups / modalities Comparison tests of 2 samples −According to scale of measurement and size
>2 groups / modalities Comparison tests of more than two samples − According to scale of measurement and size
>1 factor No blocks Model, multi-factor ANOVA type
Blocks Mixed models
Follow an evolution over time Time factor only No control Graphics only (confounding effect with weather effect)
Control Quantitative measurements
--> Longitudinal models
Qualitative measurements
--> Longitudinal models
--> Proportional tests
>1 factor Quantitative measurements Longitudinal models
Qualitative measurements Longitudinal models
Proportional test
Study the relationships between variables / indicators No causal link 2 variables Quantitative & ordinal measurements
--> Correlation tests
Nominal measurements
--> Contingency coefficients
--> Correspondence Analysis
>2 variables Concordance test
Multi-factorial analyses
Causal link 2 variables Quantitative measurements
--> Regression
Qualitative measurements
--> Proportion comparisons (chi2) −
--> Correspondence Analysis
Mixed measurement scales
--> Sample comparisons
-->2 variables Quantitative measurements
--> Multiple regression
--> Principal Component Analysis
Qualitative measurements
--> Discriminant analyses
--> Multiple correspondance analysis
Mixed measurement scales
--> Model to specify
--> Mixed Factor Analysis

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.