Dokument: Field Experiments in Behavioral Economics of Education

Titel:	Field Experiments in Behavioral Economics of Education
URL für Lesezeichen:	https://docserv.uni-duesseldorf.de/servlets/DocumentServlet?id=41039
URN (NBN):	urn:nbn:de:hbz:061-20170130-092830-2
Kollektion:	Dissertationen
Sprache:	Englisch
Dokumententyp:	Wissenschaftliche Abschlussarbeiten » Dissertation
Medientyp:	Text
Autor:	Wagner, Valentin [Autor]
Dateien:	[Dateien anzeigen] Adobe PDF [Details] 5,79 MB in einer Datei [ZIP-Datei erzeugen] Dateien vom 27.01.2017 / geändert 27.01.2017
Beitragende:	Jun.-Prof. Riener (Phd), Gerhard [Gutachter] Prof. Dr. Normann, Hans-Theo [Gutachter]
Dewey Dezimal-Klassifikation:	300 Sozialwissenschaften, Soziologie » 330 Wirtschaft
Beschreibung:	In this thesis, I conduct field experiments in schools and focus on two possibilities out of many which could change pupils' motivation in school: (i) non-monetary incentives and (ii) framing manipulations. Extrinsic financial incentives are a "natural" resource of economists to solve problems in motivation. However, this type of incentives can be very costly, not feasible for policy makers and teachers along with parents are mostly critical about "cash for grades". Chapter 2 entitled "Peers or Parents? On the Signaling Value of Rewards in School" (co-authored with Gerhard Riener) therefore aims at identifying to whom pupils want to signal their academic achievement to be able to better tailor and to increase the effectiveness of non-monetary incentives. To do so, we conducted a field experiment in high- and low-achieving schools in Germany with more than 2.000 pupils. Incentives were provided for self-improvement in a mathematical test and were either predetermined or self-selected. In the latter (the Choice Treatment), pupils could choose one out of four incentives. These non-monetary incentives were selected based on a survey conducted prior to the experiment and differed with respect to the principal target audience---peers or parents. In particular, pupils could choose between a medal, a parent-letter, a homework voucher or a surprise. To test the effectiveness of rewards on pupils' performance, academic achievements in the Choice Treatment are then compared to two predetermined treatment conditions (the Fixed-Medal and Fixed-Letter Treatments) and to pupils who were not eligible to receive a reward (the Control Group). We find that pupils with lower maths grades choose a reward with a higher signaling value to their parents while high-achieving pupils tend to choose to signal their academic achievements to their peers. We find no differences in the signaling decision by gender or school type (Vocational vs. High Schools). However, the effectiveness of the predetermined incentives on test performance differs by school type. Test performance decreases significantly for pupils in High Schools but not for pupils in Vocational Schools. In contrast, when allowing for choice over the incentive, we do not observe a decrease in pupils' performance in High Schools and moreover, an increase in pupils' willingness to prepare for the test. Insights from behavioral economics and its value for applications in policymaking have been increasingly recognized by governments in recent years. In 2010, the European Commission set up the "Framework Contract for the Provision of Behavioral Studies (FCPBS)", in 2014 the US government assembled the "Social and Behavioral Sciences Team", the World Bank officially launched its "Global Insights Initiative (GINI)" in 2015 and a number of European countries (i.e. UK, Netherlands, Germany, France and Denmark) installed specialized behavioral insights teams. Despite the increasing application in policy-making, behavioral concepts have been rarely applied to the educational sector although they constitute a promising source for motivating pupils. Chapter 3 entitled "Seeking Risk or Answering Smart? Framing in Elementary Schools" therefore tests the motivational power of framing effects on pupils' decision making in a multiple-choice test, in particular, how loss and gain framing affects the quantity and quality of decision. In a field experiment in elementary schools, 1.377 pupils were randomly assigned to one of three experimental conditions: (i) gain frame (Control Group), (ii) loss frame (Loss Treatment) and (iii) gain frame with a downward shift of the point scale (Negative Treatment). According to prospect theory (Kahneman and Tversky, 1979), individuals evaluate a loss approximately twice as much as an equal gain if they are loss averse and therefore pupils should increase their performance if they are endowed with the maximum score. On average, I find that pupils in both treatment groups answer significantly more questions correctly compared to the "traditional grading". This increase is driven by two different mechanisms. While pupils in the Loss Treatment increase significantly the quantity of answered questions---seek more risk---pupils in the Negative Treatment seem to increase the quality of answers---answer more accurately. Moreover, differentiating pupils by their initial ability shows that a downward shift of the point scale is superior to loss framing. High-performers increase performance in both treatment groups but motivation is significantly crowded out for low-performers in the Loss Treatment. Performance in a test depends on pupils' motivation but could also be influenced by the testing format. Testing formats can be broadly divided into open-ended questions and multiple-choice questions. Multiple-choice university entrance exams determine access to higher education in many countries, which is one important prerequisite for later employment possibilities. The application of this testing format is problematic if it favors answering strategies of certain groups in the population. Recent experiments have identified guessing as one reason for gender diferences in performance (Pekkarinen, 2015; Baldiga, 2014), but as promotion within the educational system should depend on actual knowledge and not on how knowledge is assessed, this poses a challenge for general multiple-choice tests. Chapter 4 entitled "Answering Strategies in Multiple-Choice Tests - Differences by School Types and Gender?" (co-authored with Gerhard Riener) investigates whether answering strategies in multiple-choice tests differ between school types, gender and school grades. We address three questions: First, whether pupils in different school types apply different answering strategies? Second, whether the gender gap in guessing---boys are typically found to guess more often than girls---exists across social strata? Third, whether the gender gap exists over all school grades? To answer the first two questions, we exploit data from the randomized field experiment in Chapter 2. Pupils in secondary schools in Germany can be differentiated by social background and intellectual ability as measured by the school type (Vocational and High Schools). Our experimental data are complemented by using aggregate data of a nationwide test with more than 780.000 participants of grades 3 to 12 to shed light on the third question. We find that pupils in High School skip more answers than their counterparts in Vocational School but that they obtain higher test scores by answering more accurately. Results on gender differences reveal a gender gap in skipping math tasks only for pupils from higher socio-economic families and only if questions are difficult. However, this gap can be closed by providing extrinsic rewards for performance suggesting that the gender gap in skipping test items could be in line with a stereotype-threat explanation. Moreover, gender differences are found in all school grades and tend to increase over the years.
Lizenz:	Urheberrechtsschutz
Fachbereich / Einrichtung:	Wirtschaftswissenschaftliche Fakultät » Volkswirtschaftslehre
Dokument erstellt am:	30.01.2017
Dateien geändert am:	30.01.2017
Promotionsantrag am:	14.10.2016
Datum der Promotion:	25.01.2017