The importance of landscape composition for pest control and crop yield: A global quantitative synthesis

Using a global structural equation model of 116 studies from 28 countries, we tested three hypotheses: The ‘natural enemy hypothesis’, that natural areas increase natural enemies and suppress pests; the ‘resource concentration hypothesis’, that simplified agriculture increases pests;  and the ‘agronomic quality hypothesis’ with a structural equation model. This repository contains the dataset and R script for the analysis that examines how landscape composition directly and indirectly affects crop yield, mediated through natural enemies, biological control, and pests, across multiple studies using a structural equation model.
, , # The importance of landscape composition for pest control and crop yield: A global quantitative synthesis

Dataset DOI: [10.5061/dryad.9ghx3ffwg](10.5061/dryad.9ghx3ffwg)

## Description

This repository contains the dataset and R scripts used in the manuscript submitted to *Ecology Letters*. The analysis explores how landscape composition directly and indirectly influences crop yield mediated through natural enemies, biological control, and pests across multiple studies using a structural equation model.

**Contents**

* **Data_SEM.csv:** Excel file containing the compiled dataset from 116 studies across 28 countries. The original data from which this dataset is derived is presented in the supplementary material from Karp et al. 2018, where the data were already standardized: [https://www.pnas.org/doi/abs/10.1073/pnas.1800042115#supplementary-materials](https://www.pnas.org/doi/abs/10.1073/pnas.1800042115#supplementary-materials)
* **SI_3_Data_Exploration.r** **:** R script that perfo...,