Intro to Statistics at IU Indy - Quiz 9 Problems
Overview
Andy Baker reviews logistics for IU Indy’s statistics course, then works through Quiz 9 problems on one-sample tests, proportions, independent samples, and paired samples. He emphasizes choosing the test from the question’s wording and data, calculating test statistics or confidence intervals, and making decisions by comparing results with critical values or checking whether zero falls inside an interval.
Key takeaways
- Select a hypothesis test from the prompt’s target and wording: means use μ, proportions use p, “different” indicates two tails, and “less than” indicates a lower-tail test.
- For tests of a mean, known population standard deviation σ calls for a Z test, while an available sample standard deviation s calls for a T test; proportion tests in these examples use Z.
- For a two-tailed test at α = 0.05, split the rejection area into 0.025 in each tail; use the 0.025 T-table column and, for a Z test, critical values of ±1.96.
- For a one-tailed test at α = 0.05, use the 0.05 T-table column or a Z critical value of 1.645, with the sign determined by whether the alternative is greater than or less than.
- Independent-samples tests compare unrelated groups, while paired-samples tests analyze matched or repeated observations using consistently defined difference scores.
- For confidence intervals of a difference, zero is the key benchmark: if zero lies inside the interval, fail to reject the null of no difference; if it lies outside, reject.
Chapters
- No quiz takes place in this class; the session focuses on practice for Quiz 9.
- Lab 7 is due that week, and Lab 8 is available; Andy Baker estimates Lab 8 takes 45 to 60 minutes.
- Quiz 8 scores improved compared with earlier quizzes, a positive sign for the end of the semester.
- The research proposal is due Sunday, and many students are considering option 4, analyzing General Social Survey data.
- Potential GSS topics include political views and how they relate to education or income.
- AI tools such as ChatGPT can help identify available variables and generate ideas, but students are responsible for doing the project work.
- Quiz 9 problems will be mixed rather than grouped by test type, so students must identify the appropriate method from each prompt.
- Up to 3 bonus points can be added to Quiz 9: acknowledge watching two assigned videos and submit notes by November 12.
- The remaining 7 of the semester’s 10 bonus points are allocated to later videos and notes for Quizzes 10 and 11.
- Words such as “average” indicate a mean parameter, μ, while “different” signals a two-tailed alternative.
- A stated or claimed population mean, such as 45 minutes, supplies the null-hypothesis value μ₀.
- Use the population standard deviation symbol σ for a Z test; when only the sample standard deviation s is available, use a T test.
- For a sample of 15 students, the mean commute is 46.8 minutes and the sample standard deviation is 10.2, compared with a claimed mean of 45.
- The two-tailed test uses 14 degrees of freedom and a critical T value of ±2.145 at α = 0.05.
- The calculated statistic is approximately T = 0.683, so the result falls in the fail-to-reject region; the sample does not provide sufficient evidence that the mean differs from 45 minutes.
- The question compares Jaguar County’s graduation rate with the statewide rate of 90%, making the hypotheses H₀: p = 0.90 and Hₐ: p ≠ 0.90.
- A sample of 400 students has a graduation proportion of 0.915; the null proportion, not the sample proportion, is used to calculate the standard error.
- The standard error is 0.015 and Z = 1.00, which is inside the ±1.96 critical bounds for a two-tailed 5% test.
- The prompt asks whether Indiana’s average manufacturing wage is below the national average of $28 per hour, so the alternative is μ < 28.
- With n = 14 and sample standard deviation s = 3.1, the procedure is a one-tailed T test with 13 degrees of freedom.
- The sample mean is $25.80 and the test statistic is about T = -2.655; it exceeds the lower-tail critical boundary in magnitude, supporting rejection of the null.
- The question tests whether average summer rainfall is less than the historical mean of 4 inches.
- Because the population standard deviation σ = 1.8 is known, the test uses Z; the sample has n = 50 and a mean of 3.45 inches.
- The standard error is approximately 0.255 and Z ≈ -2.157, below the lower-tail critical value of -1.645, so the null hypothesis is rejected.
- An independent-samples T test compares the means of two distinct groups rather than a sample mean with a single population benchmark.
- The formula uses the difference between group means, X̄₁ − X̄₂, and combines the groups’ variances to calculate the standard error.
- Andy Baker distinguishes independent groups from paired observations, such as measurements taken from the same person before and after.
- Group 1 has n = 10, mean 15, and standard deviation 3; Group 2 has n = 10, mean 18, and standard deviation 4.
- The estimated mean difference is 15 − 18 = -3, and the standard error is √(3²/10 + 4²/10) ≈ 1.581.
- The resulting T statistic is about -1.898, with 18 degrees of freedom and a two-tailed critical value of ±2.101.
- The approximate 95% confidence interval for the difference in group means is -3 ± 2(1.581), or about [-6.2, 0.2].
- Because zero lies in the interval, the result fails to reject the null hypothesis of no mean difference.
- The same decision follows from comparing T ≈ -1.898 with the ±2.101 critical values; the confidence-interval method is the one requested for this example.
- Paired samples apply to linked observations, especially before-and-after measurements on one person.
- Other examples include twins or a student’s education compared with their father’s education, where observations are related rather than independent.
- Each paired difference must follow one consistent direction, such as after minus before, throughout the calculation.
- Each d score represents one matched pair’s difference; the paired-samples example uses six differences and an average difference of about -0.17.
- The sample standard deviation of the differences is 1.169, giving a standard error of 1.169/√6 ≈ 0.477.
- The paired-samples method is a T test because it uses the sample standard deviation of the difference scores.
- Using the mean difference of -0.17 and standard error 0.477, the approximate interval is -0.17 ± 2(0.477), or [-1.124, 0.784].
- Zero falls within that interval, so the paired-samples result fails to reject the null hypothesis of no average change.
- The test statistic is about T = -0.356; with 5 degrees of freedom, the two-tailed critical values are ±2.571, leading to the same conclusion.
- Andy Baker directs students to additional online problem videos and invites questions about their research projects.
- A student begins discussing an anxiety-and-depression project using data to compare responses across two time points.
Transcript excerpt
0:01 All right. Looks like we are recording. Okay. So, let's go ahead and go through our schedule. So, as a reminder as a reminder, no quiz today. So, if you came prepared to take a quiz, I'm sorry. But we are going to go over the quiz for next time. That's all we're going to focus in on. It's just the quiz nine problems. So, it's going to be a lot of a lot of writing and so going through some problems and hopefully getting all the cases. And it's just it's just repetition. That's all it is. Repetition is the mother of
0:48 all learning, as I tell my 6-year-old tennis players. All right. Go back here. Lab seven is available. That I believe is that due this weekend? Okay. It's due this week. I think I got about half of them in. It seems like this one's I'm getting the less least amount of questions about. So, knock on wood that that one's a easier one to do. Lab eight is also available to you. Lab eight, I went and did it myself yesterday about 3:00 in the morning. And so it's it's it can be done. So, if there are questions, let me know. I've gone through it. I've done it myself. And it took 45 minutes to an hour for me. Quiz eight results. Here's the quiz
1:35 eight results from last time. The scores have improved. And so once again, you know, this is stuff that I I care about obviously as a teacher, as an educator, to see this high of a score this time of the semester when you're expecting numbers to go down, especially when we think about where we were maybe two, three weeks ago. So, this is great way to finish off the
This is the start of the transcript. Save this video free for the whole transcript, searchable and time-coded.
Related notes
Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Andy Baker.