Answering Mobile Surveys With Images: An Exploration Using a Computer Vision API

Abstract

Most mobile devices nowadays have a camera. Besides, posting and sharing images have been found as one of the most frequent and engaging Internet activities. However, to our knowledge, no research has explored the feasibility of asking respondents of online surveys to upload images to answer survey questions. The main goal of this article is to investigate the viability of asking respondents of an online opt-in panel to upload during a mobile web survey: First, a photo taken in the moment, and second, an image already saved on their smartphone. In addition, we want to test to what extent the Google Vision application programming interface (API), which can label images into categories, produces similar tags than a human coder. Overall, results from a survey conducted among millennials in Spain and Mexico ( N = 1,614) show that more than half of the respondents uploaded an image. Of those, 77.3% and 83.4%, respectively, complied with what the question asked. Moreover, respectively, 52.4% and 65.0% of the images were similarly codified by the Google Vision API and the human coder. In addition, the API codified 1,818 images in less than 5 min, whereas the human coder spent nearly 35 hours to complete the same task.

Testing the Use of Voice Input in a Smartphone Web Survey

“…Research is also focusing on the new opportunities offered by the increased use of smartphones, including the possibility to collect different types of data such as passive data (Revilla, Ochoa, & Loewe, 2017) or images (Bosch, Revilla, & Paura, 2018). In particular, some studies have considered the possibility of using voice input (VI) functions to make it easier and quicker for smartphone respondents to provide their answers to open narrative questions (Lütters, Friedrich-Freksa, & Egger, 2018;Revilla, Couper, & Ochoa, 2018).…”

Automated Visual Content Analysis (AVCA) in Communication Research: A Protocol for Large Scale Image Classification with Pre-Trained Computer Vision Models

“…The supervised machine learning techniques, even in combination with unsupervised approaches, allow the researcher to classify a large set of images into predefined categories or to train a customized model, and even offer the opportunity to combine unsupervised and supervised techniques to meet the challenges inherent to human coding procedures. Bosch et al (2019) compared the performance of Google Vision API's and manual coding and found that in the case of photos, 65% of the tags of the human coder and the API were similar. In further analyses, the study found that similar conclusions result from using either the human or the API tags.…”

Do Millennials differ in terms of survey participation?

“…In addition, making the survey experience more natural for Millennials may increase the engagement and future participation of this generation. For instance, allowing respondents to answer survey questions with images (Bosch et al, 2018).…”