What is the chances of that happening? (Statistics) Flashcards

1
Q

What is observed value?

A

Observed value is what we record.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is random error?

A

Random error is the difference between the observed value and the true value.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is an experimental unit?

A

An individual.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is sampling?

A

Sampling obtains a representative sample of the population to investigate abundance and distribution within a population. The goal of sampling is to obtain reproducible results which is not dependent on our specific sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the types of sampling?

A

There are two types of sampling. Random sampling and Non-random sampling.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the types of sampling?

A

There are two types of sampling. Random sampling and Non-random sampling.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is random sampling?

A

Random sampling ensures that all potential participants can enter a study without bias. Random sampling may lead to inaccurate data as not all participants can be selected within each group, leading to underestimates. Random sampling includes simple sampling and stratified sampling.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is simple sampling?

A

Simple sampling is when a researcher randomly selects a subset of participants.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is stratified sampling?

A

In stratified sampling, researchers divide a population into smaller strata based on shared characterisitics and randomly select within the strata to form a final sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is stratified sampling?

A

In stratified sampling, researchers divide a population into smaller strata based on shared characteristics and randomly select within the strata to form a final sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is blocked sampling?

A

Blocked sampling is used to reduce variance. We obtain samples from the strata in stratified sampling and place them into similar blocks. This is where the block is a factor like height to reduce variability within the block compared to the entire sample, We apply an experimental method within each block. This allows us to obtain a more efficient estimate.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Difference between blocked sampling and stratified sampling

A

Blocked sampling and stratified sampling is classifying participants into subgroups/strata. However, stratified sampling is random and based on pre-exisitng characterisitcs. In stratified sampling, researchers divide a population into smaller strata based on shared characteristics and randomly select within the strata to form a final sample. We can use the samples from strata in stratified sampling to use in blocks. In blocks, it is simply the groups that a researcher creates like an intervention block

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is systematic sampling?

A

Systematic sampling is when you choose participants out of the entire sample by using an nth number. Eg choosing every 5th participant.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a null hypothesis?

A

A null hypothesis (h0) is stating that no significant difference exists.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a null hypothesis?

A

A null hypothesis (h0) is stating that no significant difference exists.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is an alternative hypothesis?

A

An alternative hypothesis is stating that a significant difference exists.

16
Q

What is a Type 1 error?

A

Type 1 error occurs is when we reject the null hypothesis even though it is true.This is a false positive result.

17
Q

What is a Type 2 error?

A

Type 2 error is when we accept the null hypothesis as true but it is wrong. This is a False negative result.

18
Q

How do we obtain data from the sample?

A

We carry out the study of the same sample w/ treatment and observe for 30 days. We carry out study w/out treatment and observe over 30 days. Then we record the results.

18
Q

How do we obtain data from the sample?

A

We carry out the study of the same sample w/ treatment and observe for 30 days. We carry out study w/out treatment and observe over 30 days. Then we record the results.

19
Q

What is probability?

A

Probability is on a scale from 0 -> 1. 0 is that it is certain not to happen . 1 is certain that it will happen.

20
Q

Why do we use p-value?

A

In order to determine if differences in results are due to its effectiveness rather than chance and disprove the null hypothesis.

21
Q

What is test statistic?

A

The statistic (data) that is being assessed which we compare to the results expected of a null hypothesis. Eg, if we’re doing a coin toss and saying its a fair coin, the test statistic is the probability of the coin being on heads. We then convert this into a p-value

22
Q

How do we carry out statistical testing?

A

We state the null hypothesis and alternative hypothesis. Define and evaluate a test statistic. Convert test statistic into a p value. Calculate the p value and interpret our results.

Essentially, we are going to calculate the p-value if the null hypothesis is true and compare this to the p-value of our observed results to reject or accept the null hypothesis.

23
Q

What is the p-value?

A

A measure of significance which represents the probability of occurrence to prove or disprove the null hypothesis.

24
Q

What are odds and odds ratio?

A

Odds is the probability of an occurrence/probability of it not occurring.
Odds ratio is the ratio of odds in one group against the other.

25
Q

What is effect estimate?

A

Effect estimate describes the differences in values between the samples. Examples of effect estimate is relative risk.

26
Q

How does p-value change with a greater sample size?

A

When we increase the sample size, the test statistic doesn’t change. However, the p value will be affected as a larger sample reduces the risk of random error and is more likely to detect any significance if it exists.

27
Q

What is the 95% confidence interval?

A

The probability that the parameter will fall within a set of values. We add/subtract 1.96 from the sample mean to represent standard error.

28
Q

What is standard deviation?

A

The spread of values around the mean.

29
Q

What is the Z value?

A

The number of standard deviations a specific data point is away from the mean. We calculate this by doing (raw score- mean) and dividing this by standard deviation.

30
Q

How to calculate z value?

A

Data - mean/ standard deviation

30
Q

How to calculate z value?
Data - mean/ standard deviation

A
31
Q

How to calculate 95% confidence interval?

A

Mean +/- 1.96 x (standard deviation /total sample size)

32
Q

What happens to 95% confidence interval if standard deviation increases?

A

95% confidence interval will increase alongside this.

33
Q

Why is repeat sampling beneficial?

A

It reduces the risk of human error.