Real-Effort Tasks in Experiments: The Task Choice Matters
OPEN ACCESS
Author / Producer
Date
2021
Publication Type
Doctoral Thesis
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
Real‐effort tasks are widely used in experimental research to study effort provision. A great variety of tasks exists, each of which has its own properties. The tasks are employed in different areas of application. The tasks carry different properties, which involve advantages and disadvantages depending on the specific area they are applied. This thesis is about the choice of task and its implications. It is structured as follows.
The first chapter introduces the broader topic and provides background information on measuring effort in experiments. As real‐effort tasks greatly differ in their design, several ways of classifying them are presented. The task classification includes their degree of realism, the extent to which the output produced is useful, and the skills and character traits required to perform a particular task well.
The second chapter provides an overview of the literature that criticizes the design and implementation of real‐effort tasks. As a synthesis of the corresponding literature, a series of design criteria are presented. The criteria aim to improve experimental control while maintaining the greater realism of real‐effort measurements compared to stated effort. To achieve this, design practices are presented in order to enhance experimental control over the effort‐cost function to ensure that voluntary effort provision is kept to a minimum and over the output‐production function to ensure that actual effort is required to complete the task.
To evaluate and compare tasks with regard to these aspects, the third chapter introduces a new methodology, the real‐effort task survey. The survey is filled out by (prospective) study participants and determines their subjective perception of the task design. This is crucial because only they themselves can judge i) to what extent a task motivates them to make voluntary efforts and ii) how strenuous it is for them. Furthermore, the results of a first application of the survey are presented, comparing seven frequently used task types.
To shed light on the impact of task properties on effort measurements, Chapter four examines the influence of subject characteristics on individual performance. To this end, the study presented in Chapter three contains several additional elements to characterize the study participants. Using methods of motivation diagnostics and machine learning, abilities, personality, and motivation are found to explain a large part of the variation in the subjects’ observed effort.
Chapter five concludes this thesis, sums up the results and contributions, puts them into relation, and provides an outlook on prospective research.
To conclude, this work aims to raise awareness of the various properties of tasks, their differences, and their varying suitability for a given application. The thesis makes several conceptual and methodological contributions and serves the practitioner to classify, design, select, and implement tasks. In summary, tasks are not simply neutral and interchangeable. This is why the choice of task is vital and must match the research question being investigated.
Permanent link
Publication status
published
External links
Editor
Contributors
Examiner: Köthenbürger, Marko
Examiner: Schmid, Petra
Book title
Journal / series
Volume
1
Pages / Article No.
Publisher
ETH Zurich
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Economics; Economics, Cognition, Economics: General Interest, Psychology, Psychology: General Interest; Behavioral economics; Experimental design; Experimental economics; Real-effort tasks; Psychology; Motivational psychology
Organisational unit
03988 - Köthenbürger, Marko / Köthenbürger, Marko