PSM530 Review for PSM403

What's wrong with the hypothesis below?

H1. There is no relationship between video game use and reaction time.

• The hypothesis predicts a lack of relationship.

• The hypothesis not logical

• Nothing, this is a good hypothesis

Answer explanation

Hypotheses should always be logical, testable, refutable and positively worded.

Which of the below are types of categorical data? Select all that apply.

• Binary

• Ratio

• Nominal

• Ordinal

Answer explanation

Categorical data is made up of categories. For example:

Binary: Yes/No, Male/Female

Ordinal: 1st prize, 2nd prize, 3rd prize

Nominal: Blue/Yellow/Red, Mild/Moderate/Severe

Which of the below are types of continuous data? Select all that apply.

• Interval

• Ordinal

• Ratio

• Binary

Answer explanation

Interval data means that equal intervals on the variable represent equal differences in the property being measured. For example, temperature. The difference between 2 degrees and 4 degrees is the same as the difference between 6 degrees and 8 degrees.

Ratio data is the same, but the ratios of scores on the scale must also make sense and it has a true 0. Example: Weight. Something that weighs 12 pounds is double the weight of something that weighs 6 pounds. You can have an absence of weight (0 pounds).

The doctor in this image has concluded that there is an effect, when actually, there isn't one. What type of error is he committing?

• Type I

• Type II

Answer explanation

Type I error occurs when we believe that there is a genuine effect in our population when, in fact, there isn’t.

Type II error occurs when we believe that there is no effect in the population when, in reality, there is.

We assume normality in samples of 30+ because of the....

• Normality hypothesis

• Central Limits Theorem

• Butterfly effect

• Pearson-Spearman curve

Answer explanation

The central limits theorem states that given a random and independent samples of N observations each, the distribution of sample means approaches normality as the size of N increases

...huh?

Basically, this means that distribution of the population doesn’t matter if sample size is large. Because the sampling distribution of the mean is usually normal for large samples (n>40).

To find out if your data is normally distributed, which of the following methods could you use?

• Check skew & kurtosis numbers

• Run a spearman correlation

• Create a histogram

• Run Levene's test

Answer explanation

Histograms and skew & kurtosis checks are great ways to double check normality.

Levene's test is used to check homogeneity of variance for t-tests.

You run a Pearson's correlation between job stress and work-life balance.

You find a significant correlation of r=0.71. Is this a weak, moderate or strong correlation?

• Weak correlation

• Moderate correlation

• Strong correlation

Answer explanation

Pearson correlations are between -1 and +1.

Correlations .1 to .3 are considered weak, .4 to .6 are considered moderate and .7 to .9 are considered strong.

+1 or -1 is a perfect correlation

If you ran a t-test, you would be looking for....

• Whether the sample had a normal distribution

• The linear relationship between two variables

• Differences between groups

• Infinity stones

Answer explanation

T-tests look to see if there are significant differences between two groups! This term, we'll learn how to test for differences between 3+ groups with ANOVA.