Search

In this article, I discuss the method of relative distribution analysis and present Stata software implementing various elements of the methodology. The relative distribution is the distribution of the relative ranks that the outcomes from one distribution take on in another distribution. The methodology can be used, for example, to compare the distribution of wages between men and women. The presented software, reldist, estimates the relative cumulative distribution and the relative density, as well as the relative polarization, divergence, and other summary measures of the relative ranks. It also provides functionality such as location and shape decompositions or covariate balancing. Statistical inference is implemented in terms of influence functions and supports estimation for complex samples.

ZusammenfassungDer geschlechtsspezifische Lohnunterschied hat sich in der Schweiz ebenso wie in Deutschland in den letzten zwei Jahrzehnten zwar leicht verringert, die Lücke ist aber immer noch beträchtlich und lässt sich nur zum Teil durch produktivitätsrelevante Faktoren erklären. Um zu untersuchen, ob sich ein entsprechender „gender wage gap“ auch darin wiederfindet, welche Löhne als gerecht angesehen werden, haben wir im Rahmen von Schweizer Bevölkerungsumfragen drei randomisierte Vignettenexperimente durchgeführt. Im Unterschied zu den meisten anderen Experimenten wurde den Befragten nur jeweils eine Vignette vorgelegt, um Einflüsse sozialer Wünschbarkeit zu vermindern. Das erste Experiment belegt eine geschlechtsspezifische Doppelmoral bei der Einkommensbewertung: Bei Männern wurde ein gegebenes Einkommen eher als zu gering beurteilt als bei Frauen. Der Befund konnte in einem zweiten Experiment mit ähnlichem Design jedoch nicht repliziert werden, wobei ein zentraler Unterschied zwischen den beiden Experimenten in dem in den Vignetten beschriebenen Haushaltskontext lag. In einem dritten Experiment haben wir deshalb den Einfluss der familiären Situation systematisch untersucht. Die Ergebnisse zeigen, dass nur bei verheirateten Personen ein Unterschied zwischen Frauen und Männern gemacht wird, nicht jedoch bei Singles. Im Einklang mit dem Stereotyp des männlichen Haupternährers zeigt sich ein ausgeprägter Effekt einer „Heiratsprämie“. Verheirateten Männern wird in der Wahrnehmung der Bevölkerung bei sonst gleichen Merkmalen ein höherer Lohn zugestanden als verheirateten Frauen.

In this article, I review The Mata Book: A Book for Serious Programmers and Those Who Want to Be, by William Gould (2018, Stata Press).

In this article, I discuss the use of webdoc for creating HTML or Markdown documents from within Stata. The webdoc command provides a way to embed HTML or Markdown code directly in a do-file and automate the integration of results from Stata in the final document. The command can be used, for example, to create a webpage documenting your data analysis, including all Stata output and graphs. More generally, the command can be used to create and maintain a website that contains results computed by Stata.

Um den Fachkräftemangel im MINT-Bereich zu bekämpfen, bemühen sich Bund und Kantone, auf allen Bildungsstufen Interesse an Mathematik, Informatik, Naturwissenschaften und Technik zu wecken. Unter anderem soll dabei geschlechtsspezifischen Unterschieden hinsichtlich der Lehrstellen- bzw. Fächerwahl entgegengewirkt werden. Der vorliegende Beitrag setzt hier an und analysiert anhand der Daten der "ÜGK 2016", wie viele Jugendliche am Ende der obligatorischen Schulzeit denken, dass sie im Alter von 30 Jahren einen Beruf im Bereich der MINT-Fachkräfte ausüben werden. Dabei wird deutlich, dass bei jungen Frauen die Unterschätzung ihrer mathematischen Kompetenzen dazu beiträgt, dass sie sich seltener eine berufliche Zukunft im MINT-Bereich vorstellen können als junge Männer. (DIPF/Orig.), Pour lutter contre la pénurie de main-d’oeuvre qualifiée dans le secteur des STIM (sciences, technologies, ingénierie et mathématiques), la Confédération et les cantons s’efforcent de susciter l’intérêt pour les STIM à tous les niveaux de l’enseignement. Cette mesure vise notamment à contrecarrer les différences entre les sexes lors du choix des apprentissages et des matières. Dans ce contexte, nous utilisons les données "ÜGK 2016" pour analyser la part de jeunes en fin de scolarité obligatoire qui pensent avoir un emploi dans le secteur des STIM à l’âge de 30 ans. Les résultats montrent clairement qu’en raison d’une sous-estimation de leurs propres compétences en mathématiques, les jeunes femmes sont moins susceptibles que les jeunes hommes d’imaginer une carrière dans le domaine des STIM. (DIPF/Orig.)

At least since Thomas Piketty's best-selling "Capital in the Twenty-First Century" (2014, Cambridge, MA: The Belknap Press), percentile shares have become a popular approach for analyzing distributional inequalities. In their work on the development of top incomes, Piketty and collaborators typically report top-percentage shares, using varying percentages as thresholds (top 10%, top 1%, top 0.1%, etc.). However, analysis of percentile shares at other positions in the distribution may also be of interest. In this paper I present a new Stata command called -pshare- that estimates percentile shares from individual-level data and displays the results using histograms or stacked bar charts. A shorter version of this paper has been published in The Stata Journal 16(2): 264-300 (see: http://www.stata-journal.com/article.html?article=st0432).

In Jann (2018b, Stata Journal 18: 491–502), I presented a command called grstyle that simplifies changing the default look of Stata graphs. The command, however, still relies on idiosyncratic scheme file syntax, which may not be well known to many users. In this article, I therefore present an extension called grstyle set that automates the creation of sets of scheme file entries for a number of potentially useful adjustments, without much typing and without requiring much knowledge about scheme file syntax. Covered topics include, for example, the rendering of the background and coordinate system, the placement and look of the legend, the assignment of colors, symbols, and line patterns, and the assignment of relative or absolute sizes.

The overall look of Stata's graphs is determined by so-called scheme files. Scheme files are system components; that is, they are part of the local Stata installation. In this article, I argue that style settings deviating from default schemes should be part of the script producing the graphs, rather than being kept in separate scheme files, and I present a simple tool called grstyle that supports such practice.

The question of how cooperation and social order can evolve from a Hobbesian state of nature of a “war of all against all” has always been at the core of social scientific inquiry. Social dilemmas are the main analytical paradigm used by social scientists to explain competition, cooperation, and conflict in human groups. The formal analysis of social dilemmas allows for identifying the conditions under which cooperation evolves or unravels. This knowledge informs the design of institutions that promote cooperative behavior. Yet to gain practical relevance in policymaking and institutional design, predictions derived from the analysis of social dilemmas must be put to an empirical test. The collection of articles in this book gives an overview of state-of-the-art research on social dilemmas, institutions, and the evolution of cooperation. It covers theoretical contributions and offers a broad range of examples on how theoretical insights can be empirically verified and applied to cooperation problems in everyday life. By bringing together a group of distinguished scholars, the book fills an important gap in sociological scholarship and addresses some of the most interesting questions of human sociality.

In a seminal experiment, Doob and Gross (1968) examined the influence of social status on peer-punishment of norm violations in traffic. They observed an inverse relationship between the economic status indicated by a car that was blocking an intersection and the punishment meted out to the driver of that car, with “punishment” taking the form of a honk of the car horn. In a more recent experiment, Diekmann et al. (1996) noted the status and reactions of the cars blocked by a single midstatus car. Blocked drivers at the wheel of a higher-status car were found to punish more aggressively than drivers of a lower-status car. Our study employs a combined design to separate the effects of driver and blocker status. In two field experiments, we varied the status of the norm-violating car and recorded the status of the blocked driver’s (i.e., the experimental subject’s) car. Our results provide evidence that social distance facilitates peer-punishment. Punishment was expressed less readily when the blocked and blocking cars indicated a similar social status.

Graphical display of regression results has become increasingly popular in presentations and in scientific literature because graphs are often much easier to read than tables. Such plots can be produced in Stata by the marginsplot command (see [R] marginsplot). However, while marginsplot is versatile and flexible, it has two major limitations: it can only process results left behind by margins (see [R] margins), and it can handle only one set of results at a time. In this article, I introduce a new command called coefplot that overcomes these limitations. It plots results from any estimation command and combines results from several models into one graph. The default behavior of coefplot is to plot markers for coefficients and horizontal spikes for confidence intervals. However, coefplot can also produce other types of graphs. I illustrate the capabilities of coefplot by using a series of examples.

Umfragen uber Vorurteile fuhren oft zu verzerrten Ergebnissen, da die Befragten haufig „sozial erwunschte“ Antworten geben. In unserer Studie uber Vorurteile und Diskriminierung haben wir dagegen vier Feldexperimente in der Stadt Zurich durchgefuhrt, in denen jeweils von Angehorigen verschiedener Nationalitaten oder religioser Gruppen eine Hilfeleistung erbeten wurde. Beispielsweise wurden Passanten von einer Person mit Kopftuch um eine Hilfeleistung gebeten; in der Kontrollbedingung dagegen ohne Kopftuch. In einem anderen Experiment wurden Personen in Hochdeutsch angesprochen und in der Kontrollsituation im Schweizer Dialekt. Ein funftes Experiment bezog sich auf die Reaktionen von Arbeitgebern auf Initiativbewerbungen. Es zeigte sich, dass im Alltagsverhalten keine signifikanten Unterschiede im Ausmas der Hilfeleistung zwischen den Gruppen nachweisbar waren. Alle vier Experimente zu kleinen Hilfen im Alltag ergaben weder bezuglich Deutschen noch muslimischen Minderheiten Hinweise auf diskriminierendes Verhalten. Deutliche Hinweise gibt es dagegen fur die Diskriminierung bestimmter ethnischer Gruppen auf dem Arbeitsmarkt, auch wenn Bewerber die gleiche Qualifikation wie Schweizer und die Schweizer Staatsburgerschaft besitzen.

This article contributes to an ongoing debate about how to measure sensitive topics in population surveys. We propose a novel technique that can be applied to the measurement of quantitative sensitive variables: the item sum technique (IST). This method is closely related to the item count technique, which was developed for the measurement of dichotomous sensitive items. First, we provide a description of our new technique and discuss how data collected by the IST can be analyzed. Second, we present the results of a CATI survey on undeclared work in Germany, in which the IST has been applied. Using an experimental design, we compare the IST to direct questioning. Our empirical results indicate that the IST is a promising data-collection technique for sensitive questions. We conclude by discussing the limitations of the new technique and outlining possible improvements for future studies.

Summary This article evaluates three different questioning techniques for measuring the prevalence of plagiarism in student papers: the randomized response technique (RRT), the item count technique (ICT), and the crosswise model (CM). In three independent experimental surveys with Swiss and German university students as subjects (two web surveys and a survey using paper and- pencil questionnaires in a classroom setting), each of the three techniques is compared to direct questioning and evaluated based on the “more-is-better” assumption. According to our results the RRT and the ICT failed to reduce social desirability bias in self-reports of plagiarism. In contrast, the CM was more successful in eliciting a significantly higher rate of reported sensitive behavior than direct questioning. One reason for the success of the CM, we believe, is that it overcomes the “self-protective no” bias known from the RRT (and which may also be a potential problem in the ICT).We find rates of up to 22 percent of students who declared that they ever intentionally adopted a passage from someone else’s work without citing it. Severe plagiarism such as handing in someone else’s paper as one’s own, however, seems to be less frequent with rates of about 1 to 2 percent.

Gaining valid answers to so-called sensitive questions is an age-old problem in survey research. Various techniques have been developed to guarantee anonymity and minimize the respondent's feelings of jeopardy. Two such techniques are the randomized response technique (RRT) and the unmatched count technique (UCT). In this study we evaluate the effectiveness of different implementations of the RRT (using a forced-response design) in a computer-assisted setting and also compare the use of the RRT to that of the UCT. The techniques are evaluated according to various quality criteria, such as the prevalence estimates they provide, the ease of their use, and respondent trust in the techniques. Our results indicate that the RRTs are problematic with respect to several domains, such as the limited trust they inspire and non-response, and that the RRT estimates are unreliable due to a strong false "no" bias, especially for the more sensitive questions. The UCT, however, performed well compared to the RRTs on all the evaluated measures. The UCT estimates also had more face validity than the RRT estimates. We conclude that the UCT is a promising alternative to RRT in self-administered surveys and that future research should be directed towards evaluating and improving the technique.

Is Benford’s law a good instrument to detect fraud in reports of statistical and scientific data? For a valid test, the probability of ‘false positives’ and ‘false negatives’ has to be low. However, it is very doubtful whether the Benford distribution is an appropriate tool to discriminate between manipulated and non-manipulated estimates. Further research should focus more on the validity of the test and test results should be interpreted more carefully.

The SPost user package (Long and Freese, 2006, Regression Models for Categorical Dependent Variables Using Stata [Stata Press]) is a suite of postestimation commands to compute additional tests and effects representations for a variety of regression models. To facilitate and automate the task of tabulating results from SPost commands for inclusion in reports, publications, and presentations, we introduce tools to integrate SPost with the estout user package (Jann, 2005, Stata Journal 5: 288–308; 2007, Stata Journal 7: 227–244). The estadd command can retrieve results computed by the SPost commands brant, fitstat, listcoef, mlogtest, prchange, prvalue, and asprvalue. These results can then be tabulated by esttab or estout.

The counterfactual decomposition technique popularized by Blinder (1973, Journal of Human Resources, 436–455) and Oaxaca (1973, International Economic Review, 693–709) is widely used to study mean outcome differences between groups. For example, the technique is often used to analyze wage gaps by sex or race. This article summarizes the technique and addresses several complications, such as the identification of effects of categorical predictors in the detailed decomposition or the estimation of standard errors. A new command called oaxaca is introduced, and examples illustrating its usage are given.

At least since Thomas Piketty's best-selling "Capital in the Twenty-First Century" (2014, Cambridge, MA: The Belknap Press), percentile shares have become a popular approach for analyzing distributional inequalities. In their work on the development of top incomes, Piketty and collaborators typically report top-percentage shares, using varying percentages as thresholds (top 10%, top 1%, top 0.1%, etc.). However, analysis of percentile shares at other positions in the distribution may also be of interest. In this paper I present a new Stata command called pshare that estimates percentile shares from individual-level data and displays the results using histograms or stacked bar charts.

We analyze the changing relationship between education and assortative mating over the course of educational expansion in Switzerland between 1970 and 2000. The main question is whether educational expansion has resulted in increased openness of partnership opportunities or whether the educational system became increasingly important for assortative mating. Census data is used to describe this social change employing a cohort design. Over time and across cohorts, the proportion of people who live without a partner has increased, but the educational classes became more similar with respect to partnerlessness. At the same time, overall educational homogamy of partnerships has remained rather stable, although there were different trends for each of the educational levels. Educational expansion has contributed to increasing heterogamy for less educated and untrained persons while homogamy has increased for persons achieving intermediate and higher levels of education. However, after taking opportunity structure into account, the inclination for educational homogamy is actually more pronounced in the lower educational groups than in the higher and, in particular, the intermediate educational groups. In this respect, one can speak of a polarization of assortative mating with social closure at the lower end of the educational scale and relative openness for intermediate educational classes. This polarization, however, declined somewhat in the course of educational expansion., Für die Schweiz wird untersucht, wie sich der Zusammenhang zwischen Bildung und Partnerwahl zwischen 1970 und 2000 im Zuge der Bildungsexpansion gewandelt hat. Im Vordergrund steht die Frage, ob mit der Bildungsexpansion eine Öffnung der Heiratschancen einherging oder ob die Bedeutung des Bildungssystems für den Partnermarkt eher zugenommen hat. Für die differenzierte Beschreibung im Kohortendesign standen Volkszählungsdaten zur Verfügung. Im Zeitverlauf hat der Anteil Personen, die ohne Partner leben, zugenommen, zwischen den Bildungsschichten fand in Bezug auf Partnerlosigkeit jedoch eine gewisse Angleichung statt. Gleichzeitig beobachten wir relativ stabile Verhältnisse bezüglich der Gesamtrate an bildungshomogamen Partnerschaften, obwohl sich aufgeschlüsselt nach Bildungsniveaus unterschiedliche Trends ergeben: Die Bildungsexpansion hat einerseits zur vermehrten Heterogamie in den unteren Bildungsgruppen beigetragen, während in den mittleren und höheren Bildungsgruppen zunehmende Homogamie zu beobachten ist. In einer um die Gelegenheitsstruktur bereinigten Betrachtungsweise zeigt sich allerdings, dass die Neigung für Homogamie in den unteren Bildungsgruppen ausgeprägter ist als in den höheren und vor allem den mittleren Bildungsgruppen. In diesem Sinne kann von einer Polarisierung des Partnermarktes mit einer stärkeren Schliessung am unteren Ende der Bildungsskala und einer relativen Offenheit in den mittleren Bildungsschichten gesprochen werden, wobei das Ausmass an Polarisierung im Zuge der Bildungsexpansion eher etwas abgenommen hat., Dans le cas de la Suisse, il s’agit d’analyser l’évolution du lien entre formation et choix du partenaire entre 1970 et 2000 au cours de l’expansion du système de formation. La principale question est de savoir si l’expansion du système de formation a entraîné une augmentation des opportunités de mariage ou si le système éducatif a plutôt gagner en importance dans le choix du partenaire. La description différenciée par cohortes se base sur des données issues de recensements de la population. Au cours de la période susmentionnée, le nombre de personnes vivant sans partenaire a augmenté. Toutefois, l’écart entre les différents niveaux d’éducation s’est nettement réduit en ce qui concerne le nombre de personnes vivant sans partenaire. Parallèlement, nous observons un taux plutôt stable de partenariats caractérisés par une homogamie de formation, bien qu’une répartition parniveau de formation révèle différentes tendances: l’expansion du système de formation a, d’une part, entraîné une plus grande hétérogamie dans les groupes au niveau de formation inférieur et, d’autre part, une plus grande homogamie dans les groupes au niveau de formation moyen et élevé. Après une prise en compte des opportunités différentielles, nous observons toutefois que le penchant pour l’homogamie est plus fort dans les groupes au niveau de formation inférieur que dans ceux au niveau de formation moyen et élevé. Nous pouvons donc parler d’une polarisation du marché de la recherche de partenaire avec une plus grande fermeture sociale pour les groupes au niveau de formation inférieur et une relative ouverture pour les groupes au niveau de formation moyen. Cette polarisation a toutefois plutôt diminué au cours de l’expansion du système de formation.