Confidence Intervals Week 12

Question 1

List the limitations of Null Hypothesis Significance Testing

Answer 1

• dichotomous reject/non-reject may impede scientific progress
• high emphasis on the results of a single sample
• not enough replication
• point estimation of mean is limited and does not acknowledge large contribution of sampling error

Question 2

Why is replication important?

Answer 2

• to gain confidence in the existence size and direction of an effect
• uncover common differences amongst means
• to find sampling error

Question 3

What are confidence intervals?

Answer 3

An interval estimation of a central tendency. It is a range of plausible values for the population mean. E.g a 95% confidence interval has 95% chance of capturing the population mean.

Question 4

Why are confidence intervals used as an alternative to null hypothesis significance testing?

Answer 4

They give more information and are better than a dichotomous reject/non reject scenario.

Question 5

What units of measurement are confidence intervals used?

Answer 5

Which ever units of measurement the researcher chooses

Question 6

If support for a poll was at 43% with a margin of error of -+ 3, what would the support look like?

Answer 6

43 -+3
40%-46%
-+3 is the confidence interval, so 95% of these points will fall between that range