Dataset Information

Using social media advertisement data to monitor the gender gap in STEM: opportunities and challenges.

ABSTRACT: Boosting the number of women and girls entering careers involving STEM (Science, Technology, Engineering and Maths) is crucial to achieving gender equality, one of the UN Sustainable Development Goals. Girls and women tend to gravitate away from STEM fields at multiple stages from childhood through mid-career. The leaky pipeline is a metaphor often used to describe the loss of women in STEM and arguably other fields before reaching senior roles. Do interests expressed on social media mirror the leaky pipeline phenomenon? In this article, we collected advertisement data (reach estimates) from Facebook and Instagram disaggregated by US metros, age, gender, and interests related to STEM. We computed the Gender Gap Index (GGI) for each US metro and age group. We found that on Instagram, the GGIs for interest in Science decrease as users' age increases, suggesting that relatively there is evidence that that women, compared to men, are losing interest in STEM at older ages. In particular, we find that on Instagram, there are plausible relative trends but implausible absolute levels. Nevertheless, is this enough to conclude that online data available from Instagram mirror the leaky pipeline phenomenon? To scrutinize this, we compared the GGIs for an interest in Science with the GGIs for placebo interests unrelated to STEM. We found that the GGIs for placebo interests follow similar age patterns as the GGIs for the interest in Science across US metros. Second, we attempted to control for the time spent on the platform by computing a usage intensity gender ratio based on the difference between daily and monthly active users. This analysis showed that the usage intensity gender ratio is higher among teenagers (13-17 years) than other older age groups, suggesting that teenage girls are more engaged on the platform that teenage boys. We hypothesize that usage intensity differences, rather than inherent interest changes, might create the illusion of a leaky pipeline. Despite the previously demonstrated value and huge potential of social media advertisement data to study social phenomena, we conclude that there is little evidence that this novel data source can measure the decline in interest in STEM for young women in the USA.

SUBMITTER: Al Tamime R

PROVIDER: S-EPMC9299278 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundSchizophrenia is a disease associated with high burden, and improvement in care is necessary. Artificial intelligence (AI) has been used to diagnose several medical conditions as well as psychiatric disorders. However, this technology requires large amounts of data to be efficient. Social media data could be used to improve diagnostic capabilities.ObjectiveThe objective of our study is to analyze the current capabilities of AI to use social media data as a diagnostic tool for psychotic disorders.MethodsA systematic review of the literature was conducted using several databases (PubMed, Embase, Cochrane, PsycInfo, and IEEE Xplore) using relevant keywords to search for articles published as of November 12, 2021. We used the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) criteria to identify, select, and critically assess the quality of the relevant studies while minimizing bias. We critically analyzed the methodology of the studies to detect any bias and presented the results.ResultsAmong the 93 studies identified, 7 studies were included for analyses. The included studies presented encouraging results. Social media data could be used in several ways to care for patients with schizophrenia, including the monitoring of patients after the first episode of psychosis. We identified several limitations in the included studies, mainly lack of access to clinical diagnostic data, small sample size, and heterogeneity in study quality. We recommend using state-of-the-art natural language processing neural networks, called language models, to model social media activity. Combined with the synthetic minority oversampling technique, language models can tackle the imbalanced data set limitation, which is a necessary constraint to train unbiased classifiers. Furthermore, language models can be easily adapted to the classification task with a procedure called "fine-tuning."ConclusionsThe use of social media data for the diagnosis of psychotic disorders is promising. However, most of the included studies had significant biases; we therefore could not draw conclusions about accuracy in clinical situations. Future studies need to use more accurate methodologies to obtain unbiased results.

Dataset Information

Using social media advertisement data to monitor the gender gap in STEM: opportunities and challenges.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets