Research Methods & Stats exam 2

What is probability?

The expected relative frequency of a particular outcome

What is the multiplication rule for probability?

Two or more events are happening at the same time

What is the addition rule for probability?

Add the probability of different sets

The gambler’s fallacy refers to:

The belief that something that occured in the past will influence future outcomes

How is probability related to the normal curve?

It represents the are underneath the curve

What are the 5 steps of hypothesis testing?

1. Restate the question as a research and null hypothesis

2. Determine characteristics of a comparison distrubution

3. Determine cut off score and if it’s extreme enough to reject null

4. Determine scores on a comparison distribution

5. Decide whether to reject or accept null hypothesis

One tailed tests…

Specifiy a certain direction (increase, more) μ1 ≤ μ2 or μ1 > μ2 ; Zcrit = 1.64

Two tailed tests…

Does not specify a direction (affected, same number) μ1 = μ2 or μ1 ≠ μ2 ; Zcrit = ±1.96

Descriptive statistics refer to:

Information that describes a distribution (mean, median, mode)

Inferential statistics…

Allow you to draw conclusions about a population based on data collected from a sample

How is the Distrubution of Means created?

By repeatedly taking random samples from a population and determining the means of those samples (Z = M -μm) / σm)

What are the key points of the Central Limit Theorem?

1. Mean of dist. of means is equal to the mean of the population

2. Variance of the dist. of means is equal to the variance of the population

3. The sampling distribution of means is normally distributed if the population is or the sample size is ≥ 30

What decision error occurs when you reject the null but it was true?

Type 1 error Alpha

What decision error occurs when you fail to reject the null?

Type 2 error Beta

When do you use a z-test?

When the mean and SD is known

When do you use a one sample t-test?

When the mean is known but the SD is not

How is the variance of a population estimated?

By calculating the average of the squared difference between each data point and the population mean

The shape of a t-distrubution…

Varies in shape based on the degress of freedom

What is the formula for calculating t?

t = M - μm / Sm ; Tcrit = ± df

Power refers to:

The probability that a study will produce statistically significant results if research hypothesis is true

What is considered adequate power?

80%

What tool is used to calculate power?

G power

How can power be increased

By increasing the sample size

How is a 95% confidence interval calculated for a z-test?

M ± Zcrit (σm)

How is a 96% confidence interval caculated for a t-test?

M ± Tcrit (Sm)

What is the relationship between power and sample size?

The z score from a sample is based on the SD of a population Z = (M - μm) / σm

What are the limits of null hypothesis significance testing?

There’s only 2 options (reject/fail to reject null), results are often misinterpreted, and it gives little information about the range of true mean when Ho is rejected

What are confidence intervals and how are they calculated?

Use estimated mean to build a range of values (95% CI, p < .05) and are calculated by computing SE, finding raw scores that are X number of SEs above or below a sample mean

Within subjects design refers to:

The same subjects are tested more than once, and are exposed to all conditions

What are some issues to be considered when using a with subject design?

There's more power between subjects and possible carryover effects (counterbalancing)

Counterbalancing is best defined as:

Varying the order of conditions for participants

When do you use a paired T test?

When comparing two related samples or two measurements from the same group

What does effect size measure?

How large a difference between groups or conditions is

How do you calculate effect size for a z test?

d = μ1 - μ2 / σ (Cohen’s d)

How do you calculate effect size for a t test?

d = μ1 - μ2 / S (Cohen’s d)

When do you use a Wilcoxon test?

When two, ordinal data samples from the same group are linked to ech other

A between-subjects design is best defined as:

Participants only our early assigned to one condition or group with random assignment

When do you use an independent means test

When the population variance is unknown and interval and ratio data

How is a 95% CI calculated for an independent means test?

t = μ1 - μ2 / Sdiff (standard error of mean ; (m1 - m2) ±tcrit (Sdiff)

When do you use a Mann-Whitney U test

When the ordinal data is very skewed between two groups