Stats Exam Flashcards - Chapter 5

Using the chart (Figure 5.11) on p. 158 as a guide, which statistical test would you choose for each of the following hypotheses? 1. There will be a difference in attitude towards good dietary practice in individuals whose hobby is athletic as compared to individuals whose hobby is model aircraft assembly. 2. There will be a strong relationship between the number of car accidents an individual has been involved in and their score on a measure of anxiety. The data are not normally distributed. 3. There will be a difference in scores on a measure of depression before and after taking part in a statistics lecture. 4. The number of days annual leave an individual has per year will be related to their score on a measure of job satisfaction. You want to generate a linear equation. A. A related t-test. B. An independent t-test. C. Linear regression. D. Pearson's product moment correlation.

1. An independent t-test 2. Pearson's product moment correlation 3. A related t-test 4. Linear Regression

Match the assumptions for parametric testing with the method you would use to check for the violation of this assumption. 1. Homogeneity of variances 2. Normally distributed data 3. Absence of extreme scores or outliers. (A) Box-plots and frequencies (B) Standard Deviations (C) Histograms

1. Homogeneity of variances = Standard Deviations 2. Normally distributed data = Histograms 3. Absence of extreme scores or outliers. = Box-Plots and Frequencies.

Which of the following would you include in your report to ensure that p values were not misleading? (A) You should report the exact p-value. (B) You should report the effect size. (C) You should report the sample size. (D) A good report would include all three of the above

A good report would include all three of the above.

What is the nature of the relationship between the statistical significance and sample size? (A) A larger sample size is more likely to yield a non significant result even when the effect is large. (B) A larger sample size is more likely to yield a significant result even when the effect is small. (C) A smaller sample size is more likely to yield a significant result when the effect is small. (D) There is no relationship, this is a trick question.

A larger sample size is more likely to yield a significant result even when the effect is small.

What is a type one error? (A) A type one error is obtaining a non significant result when it should be significant. (B) A type one error is where we reject the null hypothesis when it is true. (C) A type one error is obtaining a non-significant result by sampling error alone. (D) None of the above.

A type one error is where we reject the null hypothesis when it is true.

Which of the statements below defines a type two error? (A) A type two error occurs when you fail to reject a false null hypothesis. (B) A type two error occurs when you accept a null hypothesis which is false. (C) A type two error occurs when you reject a research hypothesis which is true. (D) All of the above.

All of the above. All of the statements are true. They are different ways of expressing a type two error.

If you decided to make the critical p-value for significance 0.001 as opposed to the conventional level of 0.05, what would the consequences be? (A) There would be fewer instances when the null could be rejected. (B) You would be less likely to make a type one error. (C) You would be more likely to make a type two error. (D) All of the above.

All of the above. You would be much less likely to make a type one error as there would be a smaller chance of significant result being found by chance alone. You would also increase the probability of making a type two error as there are fewer instances when the null hypothesis can be rejected.

How can p<.05 be interpreted? (A) The probability of obtaining the data if the null hypothesis were true is less than 5%. (B) There is a 5% chance of you making a type one error. (C) There is a less than 1 in 20 probability of the result occurring by chance alone if the null hypothesis were true. (D) All of the above.

All of the above. (All of them express the concept in different ways)

Even if there is no real relationship between variables in the population, it is highly likely that you will find a relationship between variables in your randomly selected sample. (A) True (B) False

False

A one-tailed hypothesis is also known as a directional hypothesis. Which of the below is an example of a directional hypothesis? (A) In children between the ages of six and ten years, it is hypothesised that as the amount of hours spent playing with peers increases, so their vocabulary will increase. (B) In children of six to ten years, it is hypothesized that their vocabulary will not be related to the amount of hours spent playing per week with peers. (C) In children of six to ten years, it is hypothesised that there will be a significant difference in vocabulary between those spending less than ten hours a week playing, and those spending in excess of ten hours a week playing. (D) In children between the ages of six and ten years, it is hypothesised that the amount of hours spent playing with peers per week will be related to their vocabulary.

In children between the ages of six and ten years, it is hypothesised that as the amount of hours spent playing with peers increases, so their vocabulary will increase.

Loftus (1991; 1996) criticised hypothesis testing. What was his criticism? (A) Loftus' criticism was that the null hypothesis always states that there is no relationship. The actual chance of 'no' relationship at all being found is very unusual even in biological sciences. To base probability judgments on this hypothesis is misleading. (B) Loftus' criticism was that reporting descriptive statistics and confidence intervals is misleading when testing the null hypothesis. (C) Loftus' criticism was that the null hypothesis always states that there is a relationship. The actual chance of a relationship at all being found is very unusual even in biological sciences. To base probability judgments on this hypothesis is misleading. (D) Loftus' criticism was that it is difficult for researchers to test the null hypothesis when they cannot access exact p values consistently. The results reported are therefore misleading.

Loftus' criticism was that the null hypothesis always states that there is no relationship. The actual chance of 'no' relationship being found at all is very unusual, even in biological sciences. To base probability judgments on this hypothesis is misleading.

Sometimes, due to _____, we are likely to get patterns of scores in our samples that do not accurately reflect the underlying population. (A) probability (B) relationships between variables (C) sampling error (D) scattergrams

Sampling error

Complete the following sentence. The null hypothesis is: (A) the assumption there is no relationship or difference between the variables you are testing. (B) the pattern between the variables you are testing. (C) the assumption that a significant result is unlikely. (D) the assumption that there is a relationship or difference between the variables you are testing.

The assumption there is no relationship or difference between the variables you are testing.

Hypothesis: Individuals who listen to music whilst revising will achieve significantly higher exam grades than will individuals who do their revision in silence. What are the dependent and independent variables? (A) The independent variable is music and the dependent variable is silence. (B) The independent variable is silence and the dependent variable is exam grade. (C) The independent variable is revision condition and the dependent variable is exam grade. (D) The independent variable is exam grade and the dependent variable is revision condition.

The independent variable is revision condition and the dependent variable is exam grade.

What does a p-value generally tell us? (A) The p-value tells us the likelihood of our obtaining the significant result due to a relationship between our variables in the population. (B) The p-value tells us the likelihood of our obtaining the pattern of results due to variance if there is not a relationship between our variables in the population. (C) The p-value tells us the likelihood of our obtaining the pattern of results due to sampling error if there is a population. (D) The p-value tells us the likelihood of our obtaining the pattern of results due to sampling error if there is not a relationship between our variables in the population.

The p-value tells us the likelihood of our obtaining the pattern of results due to sampling error if there is not a relationship between our variables in the population.

Hypothesis: Higher levels of depression are related to higher levels of anxiety. Considering this research hypothesis, what would the null hypothesis be? (A) Individuals with higher levels of depression will have lower levels of anxiety. Any relationships otherwise observed are the result of chance alone. (B) There will not be a significant difference between those individuals who score high on depression, in comparison to those individuals who score high on anxiety. Any observed difference is the result of chance alone. (C) There is no relationship between depression and anxiety. Any observed relationship is the result of chance. (D) Individuals with lower levels of depression will have higher levels of anxiety. Any result otherwise observed is the product of chance.

There is no relationship between depression and anxiety.

Hypothesis: Individuals who listen to music whilst revising will achieve significantly higher exam grades than will individuals who do their revision in silence. Thinking about this research hypothesis, which of the below would be an appropriate null hypothesis? (A) The more music an individual listens to when they are revising, the higher their exam grade will be. In addition, the more silence an individual experiences whilst revising, the lower their exam grade will be. (B) There will be no relationship between examination grade and the amount of music or silence experienced during revision. Any observed relationship is the product of chance alone. (C) There will be no difference in exam grade between those individuals who revise whilst listening to music and those individuals who revise in silence. Any observed differences are due to chance alone. (D) Individuals who listen to music whilst revising for their exam will achieve significantly lower exam grades than will individuals who revise in silence.

There will be no difference in exam grade between those individuals who revise whilst listening to music and those individuals who revise in silence. Any observed differences are due to chance alone.

As a solution to his criticisms of null hypothesis testing, Loftus (1991;1996) suggested that researchers should make it their practice to report confidence intervals alongside their statistical tests. (A) True (B) False

True

One of the problems we face when conducting research is that when we select samples from populations, we might not get a sample that accurately reflects that population. (A) True (B) False

True

The p-value is sometimes misinterpreted. It represents the probability of a relationship occurring by chance if the null hypothesis is true. Therefore it assumes that the null is true. (A) True (B) False

True

The second criticism Loftus (1991; 1996) raised was that psychologists don't give enough consideration to the _____________ when reporting their results. (A) exact p value (B) statistical tests (C) population means (D) null hypothesis

population means