Standing Tall Pays Off: A Meta-Analysis of Height Premium
| Název práce v češtině: | Stát zpříma se vyplácí: Meta-analýza výškového prémia |
|---|---|
| Název v anglickém jazyce: | Standing Tall Pays Off: A Meta-Analysis of Height Premium |
| Klíčová slova: | výška, mzda, determinanty mzdy, Mincerova rovnice, meta-analýza, publikační selektivita, Bayesovské průměrování modelů |
| Klíčová slova anglicky: | height, wage, wage determinants, Mincer equation, meta-analysis, publication bias, Bayesian Model Averaging |
| Akademický rok vypsání: | 2021/2022 |
| Typ práce: | diplomová práce |
| Jazyk práce: | angličtina |
| Ústav: | Institut ekonomických studií (23-IES) |
| Vedoucí / školitel: | doc. PhDr. Zuzana Havránková, Ph.D. |
| Řešitel: | skrytý - zadáno vedoucím/školitelem |
| Datum přihlášení: | 29.06.2022 |
| Datum zadání: | 29.06.2022 |
| Datum a čas obhajoby: | 21.09.2023 09:00 |
| Místo konání obhajoby: | Opletalova, O206, místnost. č. 206 |
| Datum odevzdání elektronické podoby: | 01.08.2023 |
| Datum proběhlé obhajoby: | 21.09.2023 |
| Oponenti: | Mgr. Petr Polák, M.Sc., Ph.D. |
| Seznam odborné literatury |
| Andrews, I. & M. Kasy (2019): Identification of and Correction for Publication Bias. American Economic Review 109(8): pp. 2766-2794.
Balcar, J. (2012). Supply Side Wage Determinants: Overview of Empirical Literature. Review of economic perspectives, 12(4), 207-222. Bom, P. R. D. & H. Rachinger (2019): “A Kinked Meta-Regression Model for Publication Bias Correction.” Research Synthesis Methods 10(4): pp. 497-514. Goldsmith, A. H., & Veum, J. R. (2002). Wages and the Composition of Experience. Southern economic journal, 69(2), 429. Hübler, O. (2015). Height and Wages. In J. Komlos & I. R. Kelly (Eds.), Oxford Handbooks Online. Oxford University Press |
| Předběžná náplň práce v anglickém jazyce |
| Motivation:
Mincer wage equation, describing how years of schooling and potential experience impact wage, was firstly published in 1974. The benefits of schooling are indisputable regardless of years of schooling or levels of educational attainment as explained in the review by Balcar (2012). Goldsmith & Veum (2002) conclude that worker´s experience is rewarded with a comparable return. Since Mincer equation introduction, the researchers have been extending it with various features. Apart from education and experience, thousands of papers estimate e.g. the effects of particular skills, social and economic background, beauty, health or social capital of individuals on their earnings. These are among the most often contemplated wage level determinants. But what if we extend Mincer wage equation with physical characteristics of an individual – specifically his height? Generally, the academic literature concludes that the relationship between wage and height is positive and significant. In economics, this effect is called height premium. However, Hübler (2015) points out that not only height but also additional physical factors such as gender, age, weight, early-life nutrition or physical capacity might play a role as well. In my thesis, I would like to focus exclusively on height premium – its true effect and what is it driven by. There are numerous papers investigating the relationship between height and wages theoretically and empirically but to my knowledge, there is no meta-analysis dealing with this topic. Given the ubiquitous nature of wage height premium results in the academic literature, I believe there is also room for publication bias detection as the results raise some suspicions. Hypotheses: Hypothesis #1: The literature on wage height premium is affected by publication bias Hypothesis #2: After accounting for publication bias, the height premium is lower than commonly thought Hypothesis #3: Height is not the only physical factor affecting wage (gender, age, weight etc. matter as well) Methodology: At the beginning of any meta-analysis, there are studies search and data collection. Based on my keywords, I will create different combinations of search queries in Google Scholar. Relevant studies will be gathered (i.e. studies that report estimates of the effect I am interested in and also include standard errors). As there is no previous meta-analysis, whose dataset I could build upon and extend with new studies or cross-check my search with, I expect this part to take a lot of time and diligent work. Moreover, I will do snowballing to make sure all the relevant studies published within the last 3 years were indeed included. PRISMA diagram will be created. The next step will be the data collection itself. I will collect the estimates of the effects (i.e. estimates of height premium), their standard errors and also variables that account for differences in individual studies such as age range of individuals, time range of the data, country, sample size, number of citations of the study etc. This will be followed by data inspection and data cleaning i.e. checking for outliers or suspicious values, trying to identify the source of these possible mistakes and correcting them. Publication bias occurs due to the fact that sometimes studies with unintuitive or statistically insignificant results are not published (they are “filed away in a drawer”). For example in our case we can imagine it as the negative height premium estimates of several studies that are not published because such results are not in line with the author´s expectations. But as these “bad” estimates are not reported, the simple mean of the published literature is consequently biased upwards. Publication bias will be corrected via both linear (FAT-PET, fixed effects, between effects and weighting) and non-linear techniques (selection model by Andrews & Kasy (2019) or endogenous-kink model by Bom & Rachinger (2019) – they assume that publication bias is not a linear function of standard error). Heterogeneity will be examined with the help of Bayesian and frequentist model averaging. Finally, best-practice estimate will be constructed. Expected contribution: As remarked by Balcar (2012), the volume of papers discussing the relation between various characteristics of individuals and their wage levels is enormous. A considerable fraction of those papers deals with height premium. Nevertheless, no meta-analysis on this particular topic has been formed yet. Therefore, this thesis represents a valuable contribution as it quantitatively estimates the height premium effects, investigates and corrects publication bias and thus, reveals the true effect. On top of that, best-practice estimate is provided. Outline: 1. Introduction 2. Height premium - How is the effect estimated? Why is it important? What does previous academic literature say about it? 3. Data collection - How did I collect the data? What were the selection criteria? Basic summary statistics 4. Publication bias - What is publication bias? Why could it be present? Testing for publication bias 5. Why the estimates vary - Coding for heterogeneity and identifying its sources, best-practice estimate 6. Conclusion |