In priviere/PPBstats: An R package to perform analysis found within PPB programmes

Introduction {#intro-organo}

First, chose your objective (\@ref(objective-organo)), then the analyses and the experimental design based on a decision tree.

Finally see how to implement it based on the workflow and function relations (\@ref(workflow-organo)) from formated data (\@ref(data-organo)).

Analysis according to the objectives {#objective-organo}

In order to compare different germplasms on each location for selection, three main objectives dealing with PPB programmes organoleptic analyses can be :

• Understand and check consumers preferences. This can be done through hedonic analysis (section \@ref(hedonic)).

• Characterize products sensory properties. This can be done through napping analysis (section \@ref(napping)).

• Rank product according to few sensory descriptors. This can be done through rank analysis (section \@ref(rank)).

Decision tree {#decision-tree-organo}

For each analysis, a decision tree is proposed (Figure \@ref(fig:decision-tree-Compare-different-varieties-evaluated-for-selection-in-different-locations-senso)) based on the objective, the number of products and the number of tasters. For all analysis, no training is needed for tasters.

knitr::include_graphics("figures/decision-tree_Compare-different-varieties-evaluated-for-selection-in-different-locations_senso.png")

Once you have chosen your objective, analysis and experimental design, you can taste and measure ... (section \@ref(taste)).

Workflow and function relations in PPBstats regarding organoleptic analysis {#workflow-organo}

After designing the experiment and describing the data, each family of analysis is implemented by several analysis with the same workflow :

• Format the data
• Run the model
• Check the model and visualize outputs
• Compare means, format data for biplot and visualize outputs

Figure \@ref(fig:main-workflow-organo) displays the functions and their relationships. Table \@ref(tab:function-descriptions-workflow-organo) describes each of the main functions.

You can have more information for each function by typing ?function_name in your R session. Note that check_model(), mean_comparison(), biplot_data() and plot() are S3 method. Therefore, you should type ?check_model, ?mean_comparison, ?biplot_data or ?plot.PPBstats to have general features and then see in details for specific functions.

knitr::include_graphics("figures/main-functions-organo.png")

| function name | description | | --- | --- | | design_experiment | Provides experimental design for the different situations corresponding to the choosen family of analysis | | format_data_PPBstats | Check and format the data to be used in PPBstats functions | | describe_data | Describe the data set in order to choose the appropriate analysis to carry out | | hedonic | Run hedonic analysis | | napping | Run napping anaysis | | rank | Run rank analysis | | check_model | Check if the model went well | | mean_comparisons | Get mean comparisons | | biplot_data | Format multivariate results | | plot | Build ggplot objects to visualize output | Table: (#tab:function-descriptions-workflow-organo) Function description.

Data format {#data-organo}

For organoleptic analysis data must have a specific format. Each format is displayed in the corresponding analysis section.

priviere/PPBstats documentation built on May 6, 2021, 1:20 a.m.