Inferential Statistics I at Leuphana Universität | Flashcards & Summaries

Lernmaterialien für Inferential Statistics I an der Leuphana Universität

Greife auf kostenlose Karteikarten, Zusammenfassungen, Übungsaufgaben und Altklausuren für deinen Inferential Statistics I Kurs an der Leuphana Universität zu.

TESTE DEIN WISSEN

What is true for a t-test for independent samples and a t-test for the point-biserial correlation?

Lösung anzeigen
TESTE DEIN WISSEN

The t-test for independent samples and the t-test for the point-biserial correlation is the same!

Lösung ausblenden
TESTE DEIN WISSEN

How can we transform delta into f?

Lösung anzeigen
TESTE DEIN WISSEN

d = 2f

Lösung ausblenden
TESTE DEIN WISSEN

The result of a t-test is non-significant. We haven't run a a-priori sample size analysis. Under which conditions can we interpret this result?

Lösung anzeigen
TESTE DEIN WISSEN

we can only interpret a non-significant result when the test power is sufficiently large (and beta error is sufficently small). If the result was signfiicant, we could interpret it right away and calculate effect size.

Lösung ausblenden
TESTE DEIN WISSEN

What is the t-test and what is its function and how is it calculated?

Lösung anzeigen
TESTE DEIN WISSEN
  • standardised difference between two samples
  • t-test compares two means (means between two groups) and calculates whether they are significantly different
  • we divide difference in means by standard error// pooled standard deviation and get t = -4.314
Lösung ausblenden
TESTE DEIN WISSEN

what is the R code for a one-sample test with a test-value of 1.7?

Lösung anzeigen
TESTE DEIN WISSEN

t.test(dat$abino, mu = 1.7)

Lösung ausblenden
TESTE DEIN WISSEN

When do we take the Welsh T-Test?

Lösung anzeigen
TESTE DEIN WISSEN

when the variances of the two groups are not equal

Lösung ausblenden
TESTE DEIN WISSEN

What are 3 assumptions / requirements of the NHST?

Lösung anzeigen
TESTE DEIN WISSEN
  1. The variable x is interval scaled
  2. x follows a normal distribution in the population
  3. Null hypothesis: the samples are from a population with identical means and identical variance (= variance homogeneity)
Lösung ausblenden
TESTE DEIN WISSEN

What is the function of the Levene's test?

Lösung anzeigen
TESTE DEIN WISSEN
  1. tests for variance homogeneity with an F-test (H0: variance homogeneity is given)
  2. if F-test is significant, there is no variance homogeneity --> Null hypothesis is not true
Lösung ausblenden
TESTE DEIN WISSEN

What does the null hypothesis assume?

Lösung anzeigen
TESTE DEIN WISSEN
  • an observed difference in means is random / due to a sampling error
  • the experimental manipulation does not have a systematic influence
  • The mean for the distribution of differences of means “Stichprobenkennwerte-Verteilung”) – given the null hypothesis – will have the value zero — irrespective of the mean value of the population the means were drawn from


Lösung ausblenden
TESTE DEIN WISSEN

What is the empirical alpha probability?

Lösung anzeigen
TESTE DEIN WISSEN

the probability to find the observed difference in means or any larger difference in means under the H0-t-distribution

Lösung ausblenden
TESTE DEIN WISSEN

When is the result of a test significant?

Lösung anzeigen
TESTE DEIN WISSEN

when the empirical alpha-error is smaller than the significance level determined prior to running the test (mostly, significance level is 0.05). If a result is significant, we discard the null hypothesis and decide in favor of the alternative hypothesis H1

Lösung ausblenden
TESTE DEIN WISSEN

What is the problem with sample size and significance that can occur for NHST?

Lösung anzeigen
TESTE DEIN WISSEN

only the increase in sample size makes an effect significant with the SD of the difference in means and estimated population variance staying constant!

Lösung ausblenden
  • 12727 Karteikarten
  • 469 Studierende
  • 19 Lernmaterialien

Beispielhafte Karteikarten für deinen Inferential Statistics I Kurs an der Leuphana Universität - von Kommilitonen auf StudySmarter erstellt!

Q:

What is true for a t-test for independent samples and a t-test for the point-biserial correlation?

A:

The t-test for independent samples and the t-test for the point-biserial correlation is the same!

Q:

How can we transform delta into f?

A:

d = 2f

Q:

The result of a t-test is non-significant. We haven't run a a-priori sample size analysis. Under which conditions can we interpret this result?

A:

we can only interpret a non-significant result when the test power is sufficiently large (and beta error is sufficently small). If the result was signfiicant, we could interpret it right away and calculate effect size.

Q:

What is the t-test and what is its function and how is it calculated?

A:
  • standardised difference between two samples
  • t-test compares two means (means between two groups) and calculates whether they are significantly different
  • we divide difference in means by standard error// pooled standard deviation and get t = -4.314
Q:

what is the R code for a one-sample test with a test-value of 1.7?

A:

t.test(dat$abino, mu = 1.7)

Mehr Karteikarten anzeigen
Q:

When do we take the Welsh T-Test?

A:

when the variances of the two groups are not equal

Q:

What are 3 assumptions / requirements of the NHST?

A:
  1. The variable x is interval scaled
  2. x follows a normal distribution in the population
  3. Null hypothesis: the samples are from a population with identical means and identical variance (= variance homogeneity)
Q:

What is the function of the Levene's test?

A:
  1. tests for variance homogeneity with an F-test (H0: variance homogeneity is given)
  2. if F-test is significant, there is no variance homogeneity --> Null hypothesis is not true
Q:

What does the null hypothesis assume?

A:
  • an observed difference in means is random / due to a sampling error
  • the experimental manipulation does not have a systematic influence
  • The mean for the distribution of differences of means “Stichprobenkennwerte-Verteilung”) – given the null hypothesis – will have the value zero — irrespective of the mean value of the population the means were drawn from


Q:

What is the empirical alpha probability?

A:

the probability to find the observed difference in means or any larger difference in means under the H0-t-distribution

Q:

When is the result of a test significant?

A:

when the empirical alpha-error is smaller than the significance level determined prior to running the test (mostly, significance level is 0.05). If a result is significant, we discard the null hypothesis and decide in favor of the alternative hypothesis H1

Q:

What is the problem with sample size and significance that can occur for NHST?

A:

only the increase in sample size makes an effect significant with the SD of the difference in means and estimated population variance staying constant!

Inferential Statistics I

Erstelle und finde Lernmaterialien auf StudySmarter.

Greife kostenlos auf tausende geteilte Karteikarten, Zusammenfassungen, Altklausuren und mehr zu.

Jetzt loslegen

Das sind die beliebtesten StudySmarter Kurse für deinen Studiengang Inferential Statistics I an der Leuphana Universität

Für deinen Studiengang Inferential Statistics I an der Leuphana Universität gibt es bereits viele Kurse, die von deinen Kommilitonen auf StudySmarter erstellt wurden. Karteikarten, Zusammenfassungen, Altklausuren, Übungsaufgaben und mehr warten auf dich!

Das sind die beliebtesten Inferential Statistics I Kurse im gesamten StudySmarter Universum

Statistics

Babcock University

Zum Kurs
Statistics

University of Pretoria

Zum Kurs
statistics

Taif University

Zum Kurs
Statistics

Arkansas State University

Zum Kurs

Die all-in-one Lernapp für Studierende

Greife auf Millionen geteilter Lernmaterialien der StudySmarter Community zu
Kostenlos anmelden Inferential Statistics I
Erstelle Karteikarten und Zusammenfassungen mit den StudySmarter Tools
Kostenlos loslegen Inferential Statistics I