Neyman-Pearsonian Null Hypothesis Statistical Testing & Confidence Intervals

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/21

flashcard set

Earn XP

Description and Tags

Vocabulary flashcards covering the Neyman-Pearsonian framework for Null Hypothesis Statistical Testing (NHST), Type I and Type II errors, test statistics, directional and non-directional decision rules, and confidence intervals.

22 Terms

1
New cards
<p>David Bakan</p>

David Bakan

Psychologist who characterized the test of statistical significance in psychological research as an instance of a kind of essential mindlessness in the conduct of research.

2
New cards

Steps of NHST

A seven-step procedure: (1) State the population distribution of the random variable(s), (2) State the null and alternative hypotheses, (3) State the assumptions of the test, (4) Identify the test statistic and its sampling distribution, (5) Specify alpha and a decision rule, (6) Conduct the study, calculate descriptive statistics, and check assumptions, and (7) Run the test and make a decision.

3
New cards

Central Limit Theorem (in NHST)

The theorem establishing that if a parent population distribution is X∽N(Xˉˉ,XˉˉN)X \backsim N(\bar{\bar{X}}, \frac{\bar{\bar{X}}}{N}), the sampling distribution of the sample mean is XˉN∽N(XˉˉXˉ,Xˉˉ2N)\bar{X}_N \backsim N\big(\bar{\bar{X}}_{\bar{X}}, \frac{\bar{\bar{X}}^2}{N}\big), allowing researchers to deduce expected values despite lacking direct access to the parent population.

4
New cards

Research Hypothesis

A statement that describes the general aim or substantive prediction of the researcher.

5
New cards

Null Hypothesis (H0H_0)

A testable claim specifying what is expected of the parameter(s) of interest when there is no interesting effect (e.g., H0:Xˉˉ=3H_0: \bar{\bar{X}} = 3).

6
New cards

Alternative Hypothesis (H1H_1)

The complement of H0H_0 that covers the outcome where an interesting effect exists and the research hypothesis is supported (e.g., H1:Xˉˉ≠3H_1: \bar{\bar{X}} \neq 3).

7
New cards

Non-Directional Hypotheses

Hypotheses that test for any difference without specifying direction (e.g., H0:Xˉˉ=3H_0: \bar{\bar{X}} = 3 vs. H1:Xˉˉ≠3H_1: \bar{\bar{X}} \neq 3), evaluated using a two-tailed test.

8
New cards

Directional Hypotheses

Hypotheses specifying the expected direction of the parameter difference (e.g., H0:XˉˉXˉˉ3H_0: \bar{\bar{X}} \bar{\bar{X}} 3 vs. H1:Xˉˉ<3H_1: \bar{\bar{X}} < 3 or H0:XˉˉXˉˉ3H_0: \bar{\bar{X}} \bar{\bar{X}} 3 vs. H1:Xˉˉ>3H_1: \bar{\bar{X}} > 3), evaluated using a one-tailed test.

9
New cards

Type I Error

The error committed when rejecting a null hypothesis (H0H_0) that is actually true, with a long-run probability defined as Xˉˉ=P(rejecting H0XˉˉH0 True)\bar{\bar{X}} = P(\text{rejecting } H_0 \bar{\bar{X}} H_0 \text{ True}).

10
New cards

Type II Error

The error committed when retaining/failing to reject a null hypothesis (H0H_0) that is actually false, with a probability denoted as Xˉˉ=P(type II error)\bar{\bar{X}} = P(\text{type II error}).

11
New cards

Neyman-Pearson Decision Matrix

The four possible binary decision outcomes: retaining a true H0H_0 (correct), rejecting a false H0H_0 (correct), rejecting a true H0H_0 (Type I error with probability Xˉˉ\bar{\bar{X}}), or retaining a false H0H_0 (Type II error with probability Xˉˉ\bar{\bar{X}}).

<p>The four possible binary decision outcomes: retaining a true $$H_0$$ (correct), rejecting a false $$H_0$$ (correct), rejecting a true $$H_0$$ (Type I error with probability $$\bar{\bar{X}}$$), or retaining a false $$H_0$$ (Type II error with probability $$\bar{\bar{X}}$$).</p>
12
New cards

Test Statistic (statobs\text{stat}_{obs})

A statistic calculated from sample data that is used to make an inference about an unknown population parameter.

13
New cards

Null Distribution

The sampling distribution of the test statistic or sample mean under the condition that the null hypothesis (H0H_0) is true.

14
New cards

Standardized Test Statistic (zobsz_{obs})

The standardized test statistic obtained via linear transformation using the Central Limit Theorem: zobs=Xˉ−XˉˉXˉˉXˉˉN∽N(0,1)z_{obs} = \frac{\bar{X} - \bar{\bar{X}}}{\frac{\bar{\bar{X}}}{\bar{\bar{X}}{N}}} \backsim N(0, 1).

15
New cards

Observed p-value (pobsp_{obs})

The probability of obtaining a test statistic value as extreme as or more extreme than the observed statistic, calculated under the null distribution.

16
New cards

Alpha (Xˉˉ\bar{\bar{X}}) Cutoff

The pre-specified threshold probability (such as 0.050.05) chosen to define the region of values considered improbable under the null distribution, serving as the maximum acceptable Type I error rate.

17
New cards

Critical Value (statcrit\text{stat}_{crit})

The specific boundary value or threshold associated with the cutoff specified by alpha (Xˉˉ\bar{\bar{X}}), such as zcrit=Xˉˉ1.65z_{crit} = \bar{\bar{X}} 1.65 for a one-tailed test at Xˉˉ=0.05\bar{\bar{X}} = 0.05, or zcrit=Xˉˉ1.96z_{crit} = \bar{\bar{X}} 1.96 for a two-tailed test at Xˉˉ=0.05\bar{\bar{X}} = 0.05.

18
New cards

NHST Decision Rules

Standard decision criteria: reject H0H_0 if pobs<Xˉˉp_{obs} < \bar{\bar{X}} or if statobs\text{stat}_{obs} is more extreme than statcrit\text{stat}_{crit}; otherwise retain/accept H0H_0.

19
New cards

Point Estimate

A single sample statistic (e.g., Xˉ\bar{X}, s2s^2, ss) calculated to provide a single best guess for a population parameter.

20
New cards

Confidence Interval ((1−Xˉˉ)%(1-\bar{\bar{X}})\%)

An interval estimate constructed around a sample statistic such that over infinite random sampling of size NN, (1−Xˉˉ)%(1-\bar{\bar{X}})\% of the calculated intervals will capture the true population parameter.

<p>An interval estimate constructed around a sample statistic such that over infinite random sampling of size $$N$$, $$(1-\bar{\bar{X}})\%$$ of the calculated intervals will capture the true population parameter.</p>
21
New cards

Confidence Limits (UCL and LCL)

The upper and lower numerical boundaries of a two-sided confidence interval, calculated as UCL=Xˉ+∣zcritical∣×SEUCL = \bar{X} + |z_{critical}| \times SE and LCL=Xˉ−∣zcritical∣×SELCL = \bar{X} - |z_{critical}| \times SE.

22
New cards

Confidence Interval Decision Rule

The decision criterion stating that if the confidence interval built around the sample mean captures the value of Xˉˉ\bar{\bar{X}} asserted by the null hypothesis, accept/retain H0H_0; otherwise, reject H0H_0.