This session is dedicated to bridging the gap between the formal course requirements and your personal goals as a researcher. We will begin by completing a detailed Student Needs & Background Questionnaire. Your responses will directly influence how we approach the 14 weeks of instruction, allowing us to emphasize the data collection techniques or analytical methods you find most challenging or intriguing.

I. Introduction to Statistics -- Definition: Statistics is a branch of mathematics focused on organizing, analyzing, and interpreting groups of numbers. -- Purpose: It is a tool for pursuing truth and determining the likelihood that an intuition or hunch is true. -- Benefits: Learning statistics helps students read research articles, conduct their own research, and improve general reasoning and intuition. II. Two Branches of Statistical Methods -- Descriptive Statistics: Used to summarize and describe a group of numbers from a research study. -- Inferential Statistics: Used to draw conclusions and make inferences based on data that go beyond the numbers themselves (e.g., inferring things about a large group based on a small study). III. Basic Concepts: Variables, Values, and Scores -- Definitions: --- Variable: A condition or characteristic that can vary (have different values), such as stress level or height. --- Value: A number (e.g., 4) or category (e.g., female). Score: A particular person's value on a specific variable. -- Types of Variables (Levels of Measurement): --- Numeric (Quantitative): numbers indicating "how much" of something. --- Equal-interval: Differences between numbers represent equal amounts of the thing being measured (e.g., GPA). --- Rank-order (Ordinal): Numbers represent relative ranking, where the difference between ranks may not be equal (e.g., class standing). --- Categorical (Nominal): Values are names or categories rather than numbers (e.g., gender, major). IV. Frequency Tables -- Purpose: A table that shows how frequently each specific value occurs in a dataset to make patterns easier to see. -- Construction: List all possible values from lowest to highest. Mark a tally for each score next to its value. Sum the marks to find the frequency. Calculate the percentage of the total scores for each value. Grouped Frequency Tables: Used when there are too many specific values to list individually. Scores are combined into intervals (e.g., 0–4, 5–9) to simplify the visual picture. V. Histograms -- Definition: A type of bar chart used to graph frequency tables where the height of each bar represents the frequency of a value. -- Visual difference: In histograms (for numeric variables), bars are placed next to each other without spaces. For categorical variables, the graph is called a bar graph, and spaces are left between the bars. VI. Shapes of Frequency Distributions -- Distributions are described by their shape, symmetry, and tail thickness. -- Modality (Peaks): --- Unimodal: One high point. --- Bimodal: Two fairly equal high points. --- Rectangular: All values have roughly the same frequency. -- Symmetry and Skew: --- Symmetrical: Two halves of the graph look the same. --- Skewed: Lopsided with a long "tail" on one side. The direction of the skew (left or right) is determined by the side with the tail. --- Floor/Ceiling Effects: Skews caused by scores piling up at a lower limit (floor effect) or upper limit (ceiling effect). -- Kurtosis (Tail Thickness): --- Normal Curve: A bell-shaped, unimodal, symmetrical curve. --- Heavy-tailed: Thicker tails and more peaked than a normal curve. --- Light-tailed: Thinner tails and flatter than a normal curve. VII. Application in Research -- Reporting: Research articles rarely print full frequency tables or histograms; they usually describe the distribution shapes in the text. -- Software: Researchers typically use statistical software (like SPSS) to generate these tables and graphs rather than doing them by hand

- Calculating mean, median, mode - The concept of variability: range, variance, and standard deviation - Lab: Computing descriptive stats in SPSS - Properties of the normal curve - Standardization: Calculating z-scores

Data Lab Overview: Mid-Term Prep & Olfactory Perception Analysis In this week’s applied data lab we will translate statistical theory into practice by completing a comprehensive, mid-term style mock examination. In this session, you will step into the role of a data analyst investigating human olfactory perception using real-world data from the study "Dead body or yummy bacon?". Our primary objective today is to synthesize your foundational skills, focusing strictly on data cleaning, descriptive summaries, and data visualization without relying on inferential statistics. Lab Workflow & Key Tasks Data Preparation: We will begin by reviewing the provided codebook to understand our categorical variables (like gender and odor labels) and continuous sensory ratings. You will practice loading the dataset into our statistical software and identifying missing values. Specifically, you will learn how to properly document and handle empty entries marked as "N/A" so they do not skew your subsequent analyses. Demographic Summaries: Once the data is clean, you will generate tables to accurately profile the study's participants. This involves calculating the mean, median, standard deviation, minimum, and maximum for continuous demographic variables like age. Additionally, you will construct frequency tables to report both the absolute counts and relative percentages for categorical variables like participant gender. Descriptive Statistics: Next, we will summarize the core sensory metrics provided by the participants. You will build comprehensive tables detailing the mean and standard deviation for pleasantness, intensity, familiarity, and disgustingness. We will also practice generating grouped summaries, allowing you to cross-tabulate and compare average disgust ratings across different demographics and pathogen-related labels. Data Visualization: The final portion of the lab is dedicated to producing professional, publication-quality graphics equipped with proper titles, labeled axes, and legends. You will generate histograms to assess and describe the overall distribution shape of the disgust scores. Furthermore, you will construct side-by-side boxplots to visualize gender differences in disgust, and build bar charts to clearly compare the average disgustingness of the four specific odor samples. By the end of this session, you will be fully equipped to handle raw data, describe it accurately, and visualize key patterns. This hands-on exercise is designed to directly mirror the expectations, format, and rigor of your upcoming mid-term examination.

Welcome to our second applied data lab, where we will continue our mid-term preparation by exploring a new dataset focused on mental activity during sleep. In this session, you will act as a data analyst investigating how olfactory factors influence the relatively rare occurrence of chemosensory dream content. Similar to our previous lab, our primary objective is to master data cleaning, demographic summarization, and data visualization without crossing over into inferential statistics. Lab Workflow & Key Tasks Data Preparation: We will start by importing the Chemosensory_Dreams_SleepLab.xlsx file and consulting our codebook to understand variables like the Odor Awareness Scale (OAS) and the University of Pennsylvania Smell Identification Test (UPSIT). Because this data comes from a sleep laboratory environment, you will encounter missing values due to interrupted REM cycles, failure to complete sessions, or technical failures. You will practice systematically identifying and handling these "N/A" entries to ensure your dataset is primed for accurate analysis. Demographic Summaries: Once the data is prepared, you will generate descriptive summary tables to profile the continuous age variable of our participants, calculating the mean, median, standard deviation, minimum, and maximum. Furthermore, you will construct frequency tables to clearly display the absolute counts and relative percentages of participants by gender. Descriptive Statistics: Moving to the core study variables, you will build comprehensive tables reporting the mean and standard deviation for the continuous olfactory measures, specifically the OAS and UPSIT scores. Because dream content in this study is a binary outcome, you will also practice creating grouped summary tables to report the frequency (count and percentage) of chemosensory dreams across different genders and ambient odor conditions. Data Visualization: The final phase of the lab is dedicated to crafting publication-quality data visualizations with appropriately labeled axes, legends, and clear titles. You will create a histogram to visualize the overall distribution of OAS scores, applying your knowledge to accurately describe the shape of that distribution in your report. Next, you will generate side-by-side boxplots to compare olfactory acuity (UPSIT scores) between males and females, visually assessing median differences and variance. Finally, you will construct a bar chart to illustrate the absolute frequency of positive chemosensory dream reports across the three different ambient odor conditions: odorless control, vanillin, and thioglycolic acid. By completing this mock exam, you will solidify your descriptive statistical pipeline and build confidence in your ability to translate raw psychological data into compelling, visual summaries.

Random selection vs. random assignment. The concept of the sampling distribution. Probability logic in research.

The core logic: double negatives and rejecting the null. Type I and Type II errors. Statistical significance vs. practical importance. One-tailed vs. two-tailed tests. Determining critical values and rejection regions (alpha).

Covariance and standardization. Calculating Pearson’s r. Direction and strength of relationships. Visualizing relationships (scatterplots). Assumptions of linearity and normality. Coefficient of determination. Lab: Running bivariate correlations in SPSS. Partial and semi-partial correlation (controlling for third variables). Introduction to regression logic.

Logic of comparing two groups. Calculating t for unrelated groups. Homogeneity of variance (Levene's test). Calculating t for related/repeated measures. Cohen’s d and effect size. Reporting results in APA format. Lab: Independent t-tests in SPSS.

Statistical tests are vital in data analysis, allowing researchers and analysts to derive significant insights and make informed decisions. However, selecting the appropriate statistical test can be overwhelming due to data type, research objectives, and assumptions. Choosing an incorrect test can produce inaccurate outcomes and compromise the credibility of one’s research. This session offers a comprehensive manual on determining the suitable statistical test for different scenarios, assisting researchers in navigating the complexities of data analysis.