Statistics for the Social Sciences

Hi all, and welcome to the final episode of material for our statistics course! In this episode, we tackle the last of our parametric inferential tests: linear regression. Regression is about fitting a line to our data, which is a lot like doing algebra... but don't let that scare you away! When we use SPSS, the process is quite simple: the software takes our data, figures out the line that minimizes the error (or distance between the fitted line and our actual values), and then tells us just how good that line is in terms of how much it explains in our dependent variable (aka, the coefficient of determination, or R-squared). The last bits of this episode will describe how this is done in SPSS, what remains for us with SPSS Assignment #4, our final paper, and then a brief discussion of the final exam. Thank you all for the effort and time you've committed--you're this close to being fully done!

Hello folks, and welcome to one of our last stats episodes with new material! In this episode, we're going to discuss and learn about ANOVA, which is short for analysis of variance. We will use the ANOVA tests in SPSS when we are comparing more than two groups. In short, what this test allows us to do is separate out parts of our sample on factors that we think affect our dependent variable... or in other words, some way of separating our groups that doesn't split the group into two. An example: let's say I want to compare freshmen, sophomores, juniors, and seniors on their GPA--well, an ANOVA let's me do just that. The conversation will take us through a more thorough understanding of an F test, how we calculate degrees of freedom for an ANOVA, and what we can do to figure out not only effect size but how we can know which groups really are different from one another. I'll also cover more details on SPSS Assignment #3 as well as our Final Research Paper. Questions? Shoot me an email anytime!

Hey all, welcome to our Week 12 episode on t-tests! We've only got a few short weeks left of this course... and we're building upon last week by moving from inferential tests comparing a sample and a population to t-tests, which compare 2 sample groups or 1 sample group at 2 different times. For 2 totally separate groups with no overlap, we will be doing an independent samples t-test; for a single group that is being tested at 2 different times, we will be doing a dependent samples t-test (aka a paired-samples t-test). We'll discuss these tests are a brief review of our testing steps on critical values and how we know to reject the null hypothesis, and we'll wrap the conversation with understanding more about degrees of freedom, a practice worksheet you can use from Canvas, and a bit more regarding SPSS Assignment 3. Keep up the great work--you're almost there!

Hi all, and welcome to the first podcast episode focused on conducting inferential testing! Today's episode is going to spend about the first 30 minutes reviewing our hypothesis testing steps. Yes, it's a lot of covering what we already did in the last episode, but I think this is one of those areas of statistics that really benefits from additional review. After reviewing concepts like significance, critical values, and test statistics (as well as when to reject our null hypothesis), we will move into discussion of the one-sample z-test (or confusingly in SPSS, the one-sample t-test). This test is all about comparing a population and a sample, which is not something we do often in the social sciences. But when we do know about our population and we want to evaluate whether a sample of that group is different, this is the test we use. I hope the episode is easy to follow but if not email me and let me know!

Hello all--this episode is a short 10-minute instructional guide for the Initial Draft of your paper. This episode walks through the 4 sections you need in this paper with some explanation of what I'll be looking for and what you need to include in each section. I'll keep the description short since the episode just 10 minutes, but if you need any further guidance from me don't hesitate to reach out via email!

Howdy folks, and welcome to part 2 of our discussion for Week 10 on Significance and Hypothesis Testing! This episode is all about that hypothesis testing component, particularly the errors we can run into (Type 1 or Type 2, aka false positives and false negatives) and the steps for testing we must take with our hypotheses. I will say this episode is dense--we've got a lot of major concepts (like test statistics and critical values), some complicated steps to figure out (with respect to one-tailed or two-tailed hypothesis testing), and the consideration of how we do this with SPSS (where we compare p-values with our significance level). Take your time with this one, reread the chapter as needed, and email me anytime with questions or concerns on this!

Hey all, and welcome back to our discussions on Statistics for the Social Sciences! Today's episode (and the next one) are two of the most critical for this class--I've called Week 10 the Heart of this Course for a reason! In this episode, we will start with a review of probability and the normal curve. It's been a bit since we talked about that material so a short review is hopefully helpful. But then it's on to the big new concept of this week: statistical significance. Statistical significance is all about whether what we find in our sample is due to chance or whether something else is going on... or in short, it's going to give us a sense of whether our independent and dependent variable really are related to one another. How? We use hypothesis testing and the results of our testing to determine that answer. Our next episode will walk us through the steps of hypothesis testing in more depth but for this episode I would ask that you focus on getting a sense of what statistical significance and practical significance mean. And of course, let me know when you have questions on these concepts!

Howdy folks, and welcome to the last episode before the Midterm! I know, you're all very excited for that... but pump the brakes because before we get there we need to finish our discussion of probability and the normal curve. In this episode, we review our learning about those topics and connect it to the use of z-scores, which are a way of telling us how different a specific value is from the mean. Z-scores are a measure that uses standard deviation--that value we went over all the way back in week 2! We will spend time defining z-scores and then working on calculating them... and then connecting it back to probability or the chances that our data value is normal or rare. Put on your thinking caps, push through the material, and as always let me know when and if you have questions!

Hiya folks, and welcome to one of our final podcast episodes before the Midterm Exam! In this episode, we are going to focus on 3 key points: (1) language that we use when describing samples versus populations (estimates and parameters); (2) a discussion and review of the basics of probability; and (3) connecting probability and the normal curve. There's a lot of conceptual leaps at times with these topics... so take your time with this one. And focus on concepts--thinking of the normal curve as covering 100% of our outcomes (aka linking probability with the normal curve) and then the ways that the Central Limit Theorem and Law of Large Numbers help us connect our samples back to our population. Let me know what questions you have as you make your way through this episode!

Hello again everyone, and welcome to our Week 6 episode on research and statistics! In this episode, we will walk through the research process broken down here into 7 steps: creating the question, completing a literature review, designing our study, specifying our measurements of variables (as well as our hypotheses), collecting our data, analyzing that data, and finally offering an interpretation/conclusion that addresses our research question. This episode is a bit on the long side, but it will hopefully serve you well as a guide for our Initial Draft assignment and in descriptions of the structure for the Midterm Exam!

Hi everybody, and welcome to our week 5 discussion of data quality! In this episode, we've really only got two major concepts to understand: reliability (aka consistency) and validity (aka accuracy). Of course, there are different ways to figure out our reliability (test-retest, interrater reliability, etc.) and validity (content, criterion, and construct) and we can certainly use SPSS to assist us with this... calculations like Cronbach's alpha are not something we should do by hand! What's most critical for this topic: get the general gist of these terms, understand how we can test or evaluate each quality measure, and then use the last few slides to 'test' your knowledge of validity measures!

Hi all, and welcome to a one-off episode on statistics terms! This episode is not part of our normal progression through the course but I did want to provide more discussion and explanation of terms you'd noted as difficult to understand. The episode starts with central tendency (including arithmetic mean and percentile ranks), then moves to variance (including exclusive range, inclusive range, mean deviation, and standard deviation), and wraps with some terms that don't fit neatly into one category (inferential statistics, independent variables, dependent variables, unbiased estimates, scales of measurement, and operational definitions). Email me about any terms that are still difficult to understand and I'll do my best to expand on this episode!

Hey there all, and welcome back for Episode 6 on correlation! For this week and episode, we're going to cover a few initial concepts that need to be more directly discussed: variables (including independent and dependent variables) and then the levels of measurement (nominal, ordinal, interval, and ratio). Next, we'll get into the definition and looks of correlation (as they would appear on a scatterplot), tie back how correlation requires us to have a true variable with different values, learn about the Pearson correlation coefficient, and the ways we can measure correlation when our variables aren't numbers (ex: Phi coefficient, Rank biserial coefficient, etc.). We'll wrap up with discussion of why we shouldn't confuse causation with correlation through examination of spurious correlations. Throughout this episode, I'll be asking you to think about the ways your research project relates back to the concepts... so be prepared to think about the question you want to investigate!

Howdy ho everyone! Welcome to our Week 3 episode on data visualization! In this episode, we will talk all things graphing-related: which graphs are best for showing raw data (histograms), showing change over time (line graphs are great at this), providing comparisons (bar or area charts), and more. In addition, I've included a link to a video on how graphs can be used to misrepresent data or make it harder to correctly interpret--I do hope you make time to watch it as it's only a few minutes but covers some great pointers! This episode also provides tips on how to do good visualization and then walks through some examples of graphing and data transformation in SPSS. Let me know if you have questions and if you're experiencing any difficulties using the SPSS software to create graphs this week!

Hello stats course members, and welcome to episode #4! In this episode, we spend a little over 20 minutes talking about different measures of variation or the measures we can use to understand how different the values in our dataset are from one another. The statistical measures we use for understanding variation include the range, standard deviation, and variance. We will go over the definition/explanation of each term and then we will do some practice (both by hand and in SPSS) for calculating standard deviation and variance. As always, let me know of any questions or concerns and I'll get back to you as soon as I can with more clarification and guidance!

Howdy all, and welcome back! In this episode, we begin our discussion and learning about measures of central tendency which are part of what we would consider to be descriptive statistics. We are specifically focusing on the mean, median, and mode. This episode will walk through the definition of each measure, an explanation of how we calculate them, and then some practice exercise using Excel and SPSS to find the mean, median, and mode of different datasets. In addition, we'll cover a bit about the normal distribution and when it's best to use the median or the mean. Be sure to install SPSS using the SPSS Download and Licensing Instructions on the Week 2 Roadmap Page if you're in the online only version of this class and let me know if you have any difficulties with that. The episode includes a lot of practice as well as answers for different portions of those practice exercises, so do take a few minutes to listen in and get that information!

Hi all, and welcome to the second episode of our podcast on Statistics for the Social Sciences! In this episode, we're going to take a little bit of time to further discuss/investigate the syllabus but the majority of this podcast is going to focus on defining some key terms: descriptive statistics, inferential statistics, sample, and population. We'll also spend some time walking through the basics of the research process as well as the types of questions we'll look into as part of your larger research paper. This episode wraps with the overall plan for the rest of the class and a quick note about SPSS, the statistical software we'll be making good use of to do the heavy lifting of statistical calculations. As you listen to the episode, do email me ([email protected]) if you have any concerns or questions come up and I will assist you as quickly as I can!

Hello everyone and welcome to our podcast on Statistics for the Social Sciences! In this first episode, we will get an introduction to this course and in particular to the course Canvas site. Some of the most critical tools you will need to navigate this class are the Weekly Roadmap pages which are discussed early on. We will spend about 15-20 minutes on that as well as the other parts of the Canvas site such as the Syllabus, Assignments, Grades, and Announcements. This episode wraps with a few tips to (hopefully) make working through this course a big more manageable. I'm very happy to have you joining this course and look forward to 'speaking' with you for the next several weeks!