Querying Spreadsheets: An Empirical Study

TL;DR

This paper empirically evaluates various spreadsheet querying methods, demonstrating that model-driven, especially visual, approaches significantly enhance end-user productivity in retrieving information from unstructured spreadsheets.

Contribution

It provides an empirical comparison of querying models, highlighting the effectiveness of visual model-driven queries over other approaches.

Findings

Visual model-driven queries improve user productivity

Model-driven approaches outperform textual querying

End-users find visual queries easier to use

Abstract

One of the most important assets of any company is being able to easily access information on itself and on its business. In this line, it has been observed that this important information is often stored in one of the millions of spreadsheets created every year, due to simplicity in using and manipulating such an artifact. Unfortunately, in many cases it is quite difficult to retrieve the intended information from a spreadsheet: information is often stored in a huge unstructured matrix, with no care for readability or comprehensiveness. In an attempt to aid users in the task of extracting information from a spreadsheet, researchers have been working on models, languages and tools to query. In this paper we present an empirical study evaluating such proposals assessing their usage to query spreadsheets. We investigate the use of the Google Query Function, textual model-driven querying, and visual model-driven querying. To compare these different querying approaches we present an empirical study whose results show that the end-users' productivity increases when using model-driven queries, specially using its visual representation.