Stichprobentheorie
Einleitung
Einleitung Click to read ![](./images/book.png)
In der statistischen Analyse ist eine Grundgesamtheit (Population) eine Gruppe, für die wir einen Datensatz erstellen und einige Schlussfolgerungen ziehen wollen. Eine Erhebung ist ein Verfahren, mit dem wir die zu analysierenden Daten erhalten. Erhebungen können auf der Gesamtbevölkerung basieren (zensus- oder bevölkerungsbasiert), oder wir möchten eine repräsentative Teilmenge dieser Population auswählen. Diese Teilmenge wird als "Stichprobe" bezeichnet, wenn ihre Struktur dieselbe ist wie die der Grundgesamtheit. Daten aus Erhebungen, die eine Stichprobe heranziehen, werden als stichprobenbasiert bezeichnet.
Warum werden Datensätze in Form von Stichproben erhoben, anstatt die gesamte Bevölkerung zu untersuchen (Erhebungen auf der Grundlage von Volkszählungen)? Letztere sind für Zählungen und erschöpfende Untersuchungen notwendig, erfordern jedoch den Einsatz umfangreicher Ressourcen, was zu hohen Kosten führt.
Im Gegensatz dazu sind stichprobenartige Erhebungen geeignet, wenn die Bevölkerung homogen ist, da sie ein gutes Abbild der Bevölkerung darstellen. Außerdem sind sie die einzige Möglichkeit, wenn die Grundgesamtheit unendlich groß ist oder der Erhebungsprozess zu Informationszerstörungen führen kann. In jedem Fall sparen Stichproben Zeit und Kosten.
In der Praxis haben wir in der Regel nicht die Ressourcen, um bevölkerungsbezogene Studien durchzuführen, so dass die Alternative darin besteht, unsere Analyse auf Stichproben zu stützen. Wenn wir unsere Schlussfolgerungen auf Stichprobendaten stützen, bedeutet dies, dass es eine inhärente Fehlerspanne gibt, auf die sich mehrere Faktoren auswirken können.
Die Fehlermarge wird im Wesentlichen von drei Faktoren abhängen:
- Wie homogen die Daten in der Grundgesamtheit sind: Je heterogener die Daten sind, desto größer ist die Fehlermarge, wenn alle anderen Faktoren gleich bleiben.
- Der Stichprobenumfang: Je kleiner der Umfang, desto größer ist die Fehlermarge, wenn alles andere gleich bleibt.
- Das Stichprobenverfahren: abhängig von den Merkmalen Ihrer Daten
Bei (a) können wir nicht viel tun, aber bei den Punkten (b) und (c) gibt es einen gewissen Handlungsspielraum. Zu Punkt (c) über die angewandte Stichprobentechnik ist anzumerken, dass es eine große Vielfalt an verfügbaren Stichprobentechniken gibt, die wir anwenden können. Das nachstehende Diagramm zeigt diese Vielfalt in visueller Form:
![](data:image/png;base64,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)
Wir können die Fehlermarge unserer Schlussfolgerungen nur kontrollieren, wenn wir mit Zufallsstichproben arbeiten, und die am häufigsten verwendeten Stichprobenverfahren sind die einfache Zufallsstichprobe und die geschichtete Zufallsstichprobe.
Probenahmetechniken
Einfache Zufallsstichprobe Click to read ![](./images/book.png)
Die einfache Zufallsstichprobe ist das elementarste Stichprobenverfahren, das auf einer Zufallsauswahl der untersuchten Beobachtungen beruht. Sie besteht darin, ausgehend von einer Auflistung der Einheiten der Grundgesamtheit, n dieser Einheiten zufällig auszuwählen. Aber selbst bei dieser einfachen Technik können einige Besonderheiten des Zufallsauswahlverfahrens festgelegt werden. So kann beispielsweise entschieden werden, ob die Stichprobe mit oder ohne Zurücklegen durchgeführt werden soll. Wird die Stichprobe mit Zurücklegen durchgeführt, bedeutet dies, dass jede Einheit, die nach dem Zufallsprinzip als Teil der Stichprobe ausgewählt wurde, nach jeder Ziehung wieder in die Grundgesamtheit aufgenommen wird. Dies bedeutet natürlich, dass eine Einheit mehr als einmal in die Stichprobe aufgenommen werden kann, aber es garantiert, dass die Bedingungen, unter denen die einzelnen Ziehungen stattfinden, gleich und konstant sind und die Ergebnisse der einzelnen Ziehungen unabhängig voneinander sind.
Im Gegensatz dazu wird bei einer einfachen Zufallsstichprobe ohne Zurücklegen jede Einheit nur einmal beprobt, aber wir können nicht garantieren, dass die Bedingungen entlang der Auswahlziehungen konstant sind. Stichproben mit und ohne Zurücklegen können bei kleinen Populationen zu sehr unterschiedlichen Ergebnissen führen. Sie sind nur dann gleichwertig, wenn die Größe der Grundgesamtheit (N) sehr groß ist.
Geschichtete Zufallsstichprobe Click to read ![](./images/book.png)
In manchen Fällen werden die Beobachtungen natürlich auf der Grundlage gemeinsamer Merkmale gruppiert. So werden beispielsweise Daten über die Lohnverteilung nach dem Wirtschaftszweig der Arbeitnehmer, ihrem Geschlecht oder ihrer Wohnregion gruppiert. Schichten werden als Teile der relevanten Population definiert, die eine hohe interne Homogenität aufweisen, auch wenn zwischen den Schichten eine große Variabilität besteht. Die geschichtete Stichprobe nutzt diese Gruppierung der Beobachtungen und wählt nach dem Zufallsprinzip eine Anzahl von Einheiten in jeder Schicht L (nL ) aus, so dass sich der Gesamtstichprobenumfang durch Addition der in jeder Schicht beprobten Elemente ergibt. Es gibt mehrere Kriterien für die Aufteilung des Gesamtstichprobenumfangs auf die Schichten, von denen die folgenden die häufigsten sind:
- Disproportional: gleiche Stichprobengröße in jeder Schicht
- Proportional: der Anteil der Stichprobenmitglieder entspricht dem Anteil der Bevölkerungsmitglieder in jeder Schicht
- Optimal: proportional zur Größe und Heterogenität (Varianz) der einzelnen Schichten
Unter den gleichen Bedingungen und mit den gleichen Anforderungen an Präzision und Konfidenzintervall können wir bestätigen, dass geschichtete Stichproben im Allgemeinen einen geringeren Stichprobenumfang erfordern als einfache Stichproben. Auf Fragen im Zusammenhang mit der Berechnung des Stichprobenumfangs wird im nächsten Punkt eingegangen.
Berechnung des optimalen stichprobenumfangs
Berechnung des optimalen stichprobenumfangs Click to read ![](./images/book.png)
Die goldene Regel für den Zusammenhang zwischen dem Stichprobenumfang und der Genauigkeit unserer Schätzungen lautet, dass die Fehlermarge umso geringer ist, je größer der Stichprobenumfang ist, wenn alle anderen Faktoren gleich bleiben. Die Beschaffung statistischer Daten, selbst wenn sie in Form einer Stichprobe erfolgt, kann jedoch kostspielig sein, und manchmal fehlen uns die Mittel für große Stichproben. Folglich gibt es eine Kompromisslösung, die den optimalen (minimalen) Stichprobenumfang festlegt, den wir angesichts unserer Anforderungen an die Genauigkeit (Fehlermarge) und das Vertrauen in unsere Schätzungen sowie die Heterogenität (Varianz) der relevanten Variablen in der Grundgesamtheit benötigen.
Regel: Je größer der Stichprobenumfang, desto genauer die Schätzung
|
|
![](data:image/png;base64,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) |
|
Lösung für eine einfache Stichproboben Click to read ![](./images/book.png)
Nehmen wir zunächst an, dass wir mit unserer Stichprobe den Mittelwert der Grundgesamtheit für eine kontinuierliche Variable schätzen wollen und dass unsere Stichprobe anhand einer einfachen Zufallsstichprobe ausgewählt wird. Die Formeln, die wir anwenden müssen, sind die folgenden:
Die Konstante 𝑘 stammt aus einer Normalverteilung und wird größer, wenn wir das gewünschte Konfidenzintervall erhöhen, und das Symbol 𝑒 steht für die Fehlermarge, die wir anzunehmen bereit sind. Zusätzlich müssen wir eine Annahme über die Homogenität der Variablen in der Grundgesamtheit treffen. Dies bedeutet, dass wir einen realistischen Wert (der in der Regel aus einer früheren Studie stammt) für die Varianz der Population σ 2 festlegen müssen..
In diesen Gleichungen ist n∗ die Lösung für eine einfache Zufallsstichprobe mit Zurücklegen, 𝑛 ist die Lösung für eine einfache Zufallsstichprobe ohne Zurücklegen und N ist die Größe der Grundgesamtheit. Generell gilt , und beide Lösungen konvergieren, wenn N sehr groß ist.
Wenn wir daran interessiert sind, den Anteil (P) der Einheiten in einer Grundgesamtheit zu schätzen, die ein bestimmtes Merkmal aufweisen, lauten die Ausdrücke, die erforderlich sind, um den optimalen Stichprobenumfang für dieses Stichprobenverfahren zu finden, wie folgt:
Auch hier stammt die Konstante 𝑘 aus einer Normalverteilung und wird größer, wenn wir das gewünschte Konfidenzintervall erhöhen, und der Term 𝑒 steht für die Fehlermarge, die wir bereit sind, anzunehmen. In diesem Fall müssen wir eine Annahme über den Wert von P*(1-P) machen, der die Varianz einer binären (ja/nein) Variable ist. Die übliche Lösung besteht darin, P=1-P=0,5 anzunehmen, so dass P*(1-P)=0,25 seinen Maximalwert annimmt.
Wir können diese Technik anhand eines praktischen Beispiels veranschaulichen, wie Stichprobengrößen bestimmt werden und wie die Anwendung von R uns dabei helfen kann: Der Public Broadcasting Service (PBS) in den USA schätzt regelmäßig den Prozentsatz der Bürger:innen, welche die Arbeit des US-amerikanischen Präsidenten gutheißen oder ablehnen. Im Fall von US-Präsident Joe Biden werden diese Umfragen seit Januar 2021 durchgeführt. Die folgende Grafik zeigt die Entwicklung der Schätzungen:
Bei einer kürzlich durchgeführten Umfrage in dieser Reihe wollte PBS Schätzungen mit einem Konfidenzniveau von 99 % erhalten, war bereit, eine Fehlermarge von ±4,4 % in Kauf zu nehmen, ging vom ungünstigsten Fall aus (übliche Lösung) und nahm an, dass der Prozentsatz der Befürworter (P) gleich dem Prozentsatz der Nichtbefürworter (1-P) ist. Wie viele Bürger:innen müssten unter diesen Bedingungen in die Stichprobe aufgenommen werden? Die oben dargestellten Gleichungen können in der Sprache R implementiert werden, um eine Lösung zu finden.
Zunächst müssen wir die erforderlichen Pakete installieren und laden:
![](https://datascience-project.eu/private-data/ckeditor/plugins/imageuploader/uploads/220a2e74c.jpg)
Später können wir diesen optimalen Stichprobenumfang durch Aufruf der Funktion "sample.size.prop" im Paket ermitteln. Diese Funktion ermöglicht eine Stichprobenziehung mit oder ohne Zurücklegen, obwohl es in der Praxis keine Unterschiede zwischen den Lösungen dieser beiden Alternativen gibt, da die Grundgesamtheit (N), aus der die Stichproben gezogen werden, sehr groß ist (wir können willkürlich annehmen, dass N=200.000.000). Die folgenden Codeteile berechnen die jeweiligen Lösungen für eine Stichprobe ohne und mit Zurücklegen:
Die in beiden Fällen als Lösung eine Stichprobengröße von etwa 1.000 Einheiten findet.
Lösung für geschichtete Stichproben Click to read ![](./images/book.png)
In diesem Abschnitt werden die Formeln für die Berechnung des Stichprobenumfangs bei geschichteten Stichproben ausführlich erläutert. Der Einfachheit und Klarheit halber konzentrieren wir uns nur auf den Fall der Schätzung eines Populationsmittelwerts und bieten die beiden häufigsten Lösungen an, die den Fällen der proportionalen (1) und der optimalen Verteilung (2) entsprechen:
![](https://datascience-project.eu/private-data/ckeditor/plugins/imageuploader/uploads/123e0b854.jpg) |
Wie bereits erwähnt, entspricht die Formel in beiden Fällen der Schätzung des Populationsmittelwerts für eine kontinuierliche Variable mit einer geschichteten Stichprobe ohne Zurücklegen. In diesen Ausdrücken steht Nj ür die Größe der Schicht j und σj2 für die Varianz der Variablen in dieser Schicht.
Ähnlich wie bei den Lösungen für einfache Zufallsstichproben können wir anhand eines praktischen Beispiels in der Sprache R veranschaulichen, wie der optimale Stichprobenumfang bei geschichteten Stichproben berechnet wird.
Angenommen, ein Wohlfahrtsverband führt eine Stichprobenerhebung durch, um die jährlichen Spenden seiner Mitglieder zu untersuchen, die je nach Alter in drei verschiedene Gruppen mit jeweils 100, 700 und 200 Mitgliedern eingeteilt sind. Aus einer Pilotstudie weiß die Wohltätigkeitsorganisation, dass die jeweiligen Standardabweichungen (σj) bei den jährlichen Spenden in jeder Gruppe 6 €, 30 € und 12 € betragen. Es soll der Mindeststichprobenumfang ermittelt werden, der erforderlich ist, um die mittlere jährliche Spende zu schätzen, wobei eine Fehlermarge von 2 € und ein Konfidenzniveau von 95 % angesetzt werden.
Die folgenden Codezeilen berechnen den optimalen Stichprobenumfang und bieten die Lösungen für den Fall einer proportionalen und optimalen Verteilung, indem sie die Funktion "stratasize" aus dem Paket "samplingbook" in R aufrufen:
![](https://datascience-project.eu/private-data/ckeditor/plugins/imageuploader/uploads/124b18a55.png) |
Die entsprechenden Lösungen sind 390 und 339 Einheiten, wie im Folgenden beschrieben:
Schließlich können wir uns fragen, welcher dieser beiden Stichprobenumfänge auf die Schichten aufgeteilt werden soll. Dies kann durch den Aufruf der Funktion "stratasamp" im selben Paket erfolgen:
![](https://datascience-project.eu/private-data/ckeditor/plugins/imageuploader/uploads/221d5176d.jpg)
Die Lösungen dazu lauten:
![](https://datascience-project.eu/private-data/ckeditor/plugins/imageuploader/uploads/222b81be1.jpg)
|