Flower strips have become a prevalent measure in agricultural landscapes to counteract biodiversity loss and especially promote pollinators. Although their benefits for pollinating insects have been frequently evaluated and reported, generalized conclusions about optimal settings for effective flower strips are still difficult. From the perspective of pollinators, flower strips vary distinctly in habitat quality, and the same applies for the control sites selected for scientific studies.

In this study, we used a meta-analytic approach based on a systematic review of recent studies (2009-2020) to analyze the relationship between flower strip effectiveness for pollinators and the contrast in habitat quality between flower strips and control sites. We extracted 350 data entries from 29 out of 172 studies based on available data for richness or abundance of the pollinator taxa groups Apiformes, Lepidoptera and Syrphidae as response variables, for both flower strips and control treatments. All flower strips and control treatments were assigned a habitat quality score including information on spatial dimension, floral resources and management. Moreover, we included information on landscape complexity as measured by percent cover of semi-natural habitats in the studied landscape.

In general, our results of meta-analytical models showed an increasing effect size of flower strips on pollinators for higher contrasts in habitat quality between flower strips and control treatments. This relationship was consistent across pollinator taxa and different levels of landscape complexity. Altogether, in terms of pollinator habitat quality, high-quality flower strips were more attractive than low-quality flower strips, and the reported effectiveness of flower strips decreased from low-quality to high-quality control treatments.

We recommend that results of future studies evaluating flower strips for pollinators are always linked with the contrast in habitat quality between selected flower strips and control treatments.

This dataset was collected from a systematic literature review based on a search in ISI Web of Science (WoS) Core collection and Elsevier Scopus databases. In total, 350 data entries from 29 out of 172 studies were extracted based on screening titles, abstracts, and full texts. Each data entry corresponds to a paired result on species richness (richness data) or pollinator abundance (abundance data) as compared between treatments (flower strip – control) per publication source. Data was gathered from the main text, tables and appendixes of publications or extracted from graphical results according to the details in the published article Pérez-Sánchez et al. (2023).

The dataset includes csv files (richness_data.csv and abundance_data.csv together with data_legend.txt). Missing values due to missing data from literature sources are labeled as "NA".