Notes for Session: Project One Milestones, Science Communication, Mediation/Moderation, and Experimental Design
Course Session Notes: Project One, Science Communication, and Research Design
Purpose of today’s class
Align on upcoming assignments and group work for Project One
Discuss science communication and critical analysis of research as it appears in media
Use the podcast on Trump, Tylenol, and autism as a case study
Preview exam concepts and plan targeted review based on student input
Podcast-based case study used in discussion
Focus: how scientific claims are made in public/media, strengths and limitations of those claims
Emphasis on applying research methods and critical analysis to everyday information
Ethical and public health implications discussed (e.g., Tylenol use in pregnancy and autism risk)
Exam review planning
Instructor will cover concepts students most want help with
Explicit review of challenging concepts before the exam next week
Annotated bibliography assignment (Project One, individual component)
Posted on Canvas; due Wednesday, October 8
First individual writing-intensive component of Project One
Purpose: build skills in interpreting and reading empirical articles; connect to introduction for the research proposal
Process and purpose
The introduction for the group’s proposal starts broad, reviews literature, ends with the research question and study names
All group members share the same research question and hypotheses; each introduction is independently written
Tools and resources to use
SHU library, PsycINFO, and/or Google Scholar
Identify five articles relevant to the group’s research question
Article types: 1–2 should be review/theoretical papers or meta-analyses; the remaining 3–4 should be empirical papers (original data or new analysis)
Publication recency: at least three articles published in the last 10 years; ideally all in last 5 years (to reflect current trends, esp. social media frame)
Diversity of authors: at least three different author groups
For each article: include an APA 7 style citation and a descriptive paragraph
Two sets of guiding questions depending on article type (empirical/meta-analysis vs. review/theoretical)
Paragraph should connect the article to the group’s research question, discuss potential moderator variables, similar designs, and population of interest
Top-of-assignment elements
Include research question/topic and a paragraph about the literature review process (search engines used, search terms, difficulties)
Sample model provided by instructor (for guidance)
Example topic: “What is the relationship between parenthood status and happiness?”
Include a checklist, search method description, and examples of both an empirical article and a review article
Grading (out of 20 points)
2% of final grade; 20 points total
Breakdown (approximately):
3 points per article total: 1 point for citation, 2 points for the descriptive paragraph (covering required elements)
2 points for checklist compliance (top-of-assignment checklist and itemized compliance)
3 points for including a descriptive paragraph
Additional assignment components
At the top of the assignment, include the research question/topic and a paragraph about the literature review process (search engines/terms/difficulty)
Sample inclusion: one empirical article and one review article included as examples
Questions and support
Instructor invites questions about the annotation assignment and model
Timeline and big-picture view of Project One
Research question assignment due Friday, October 3
Annotated bibliography due Wednesday, October 8
Study design assignment due Monday, October 27
Study design workshops: informal presentations of study design; two groups per day; ~30 minutes per group; discuss design details, extraneous variables, ethics
Individual introduction draft due after study design workshops
Project Two (shorter, less intensive) in November: two articles and formal slides presentation
Rationale for two-project structure: balance collaborative design work with independent literature interpretation
In-class group activity: planning and review focus
Students asked to list top 3–5 topics to review today (time permitting)
Breakouts to discuss carryover effects, moderation vs. mediation, extraneous variables vs. confounds, single-factor vs. multi-factor designs, etc.
Groups share key topics and plan to bring helpful materials to class
Science communication and the podcast discussion (in-class activity)
Acknowledgement of personal relevance and potential controversy around autism and pregnancy-related media claims
Goals of discussion: connections between podcast content and class concepts (pseudoscience, psychological myths, study designs, variables)
Key topics raised from student contributions
Extraneous variables and confounds (e.g., maternal age as a potential confound in Tylenol-autism associations)
Genetics as a control strategy: twin and sibling studies reduce genetic confounding; when controlled, association with Tylenol and autism often disappears
Distinction between correlation/association and causation; public health implications of incorrect causal claims
Ethical considerations in public messaging and risk-benefit analysis (fever management in pregnancy; Tylenol as a fever reducer)
The role of authority figures in disseminating scientific claims; importance of source credibility and context
Implications of misinterpreted science for public behavior (e.g., doctors needing to correct misinformation for patients)
Stigma and evolving understanding of autism; diagnostic criteria and gender biases in diagnosis
The burden of misinformation on public health agencies (e.g., CDC) and the need for clear communication
Key takeaways about science literacy and public interpretation
People often encounter headlines and sound bites; need context, sample details, and methodological information to interpret findings accurately
The value of going to the original source and understanding the scientists’ own claims
Detailed discussion points from the podcast (selected highlights)
Extraneous variables: broad factors that could influence outcomes (e.g., environmental exposure, pollution, genetics, age of birth parent)
Confounds: extraneous variables that vary systematically with the independent variable and threaten internal validity
Genetics in autism research: twin and sibling studies help isolate genetic factors; many apparent associations disappear when genetics are controlled
Correlation vs. causation: many studies show association; causation requires ruling out alternative explanations and confounds
Ethical implications: balancing risk and benefits in pregnancy-related decisions; potential harms of misinformation on public health behavior
Public health communication: the duty to provide accurate information and the challenge of addressing misinformation in a timely manner
Stigma and autism: evolving understanding and the importance of recognizing a spectrum of experiences
Mediation and Moderation: core definitions and in-class practice overview
Purpose: clarify how to think about mechanisms (mediation) and conditional effects (moderation)
Mediation (why does X affect Y through M)
Conceptual path: X -> M -> Y
How to test conceptually: M is changed by X, and M then changes Y
Common formulation (simple mediation):
M = aX + e_M
Y = c'X + bM + e_Y
Total effect: c = c' + ab
Moderation (for whom or under what conditions does X affect Y)
Moderator Z changes the strength/direction of the X-Y relationship
Interaction model formulation:
Y = b1 X + b2 Z + b_3 XZ + e
If XZ interaction is significant, the effect of X on Y varies by level of Z
In-class exercise with social media time and life satisfaction
Mediators proposed: sleep, procrastination, time spent with others, exercise, adaptive coping strategies
Moderators proposed: age, type of social media, content type, influencer status
Example interpretations:
Sleep as mediator: more time on social media → less sleep → lower life satisfaction
Content type as moderator: uplifting content may buffer or reverse negative effects for some users
Age as moderator: younger users may experience stronger negative effects than older users due to differing self-esteem dynamics
Moderated mediation (brief mention; not required for current projects)
Some studies examine whether the mediation pathway (X -> M -> Y) varies across levels of a moderator (Z)
Could involve a combination of mediation and moderation in a single model
For today’s coursework, students will choose either a mediator or a moderator for their project introduction; moderated mediation is optional and more advanced
Carryover effects and order effects in experimental design
Within-subjects design: participants experience multiple conditions
Carryover/order effects: performance in one condition influences performance in subsequent conditions
Counterbalancing: a method to control for order effects in within-subjects designs
Between-subjects design: different participants in each condition; random assignment helps equalize participant-level variables across conditions
Internal validity: the degree to which observed effects can be attributed to the manipulated independent variable rather than confounds or order effects
Quick recap of key terms
Within-subject design vs Between-subject design
Carryover effects
Counterbalancing (within-subjects)
Random assignment (between-subjects)
How these practices strengthen internal validity
Practical considerations for your own projects (recap and planning tips)
Decide early whether your study design will use mediation or moderation (or the more complex moderated mediation as an advanced option)
When identifying potential mediators or moderators, ensure the proposed mediator can realistically be influenced by the independent variable (i.e., X should plausibly change M; moderators are factors that differentiate effects across groups or conditions)
Consider extraneous variables and confounds early
Distinguish broad extraneous variables from confounds that vary with the independent variable
Use conceptual examples such as age, genetics, health status, environment, socioeconomic status to think through potential confounds
Ethical implications and public health relevance
Recognize the impact of misleading causal claims in the real world (e.g., pregnancy-related medication guidance)
Balance discussion of risks and benefits; acknowledge limitations and the evolving nature of scientific evidence
Use of sources and evidence quality
Include a mix of empirical and theoretical/review papers
Seek diversity of author groups; prioritize recent work (last 5–10 years, ideally)
Group work logistics and assessment
Expect to meet outside class and coordinate communication within groups
Canvas group evaluations due October 6 to balance contributions (confidential)
Key dates to remember
Research question assignment due: Friday, October 3
Annotated bibliography due: Wednesday, October 8
Study design assignment due: Monday, October 27
Exam: Tuesday (date not specified in transcript; plan accordingly)
Group evaluations due: October 6
Introduction draft due: after study design workshops
Quick glossary of terms (for study use)
Annotated bibliography: a concise summary and evaluation of each source, tied to the research question
Meta-analysis: a quantitative synthesis combining results from multiple studies
Empirical paper: a paper presenting original data or new analyses
Theoretical paper: a paper proposing new theories or frameworks without new empirical data
Moderator: a variable that changes the strength/direction of the X-Y relationship
Mediator: a variable that explains the mechanism through which X affects Y
Confound: an extraneous variable that biases the observed X-Y relationship by varying with X
Extraneous variable: any variable other than the IV that could influence the DV
Internal validity: the degree to which an experiment establishes a causal relationship between IV and DV
Carryover effect: an order effect where one condition’s effect carries over into another in a within-subject design
Counterbalancing: balancing the order of conditions across participants to mitigate order effects
Random assignment: assigning participants to conditions by chance to equalize preexisting differences
Moderated mediation: a combined model where a moderator affects the strength of mediation paths (advanced topic)
Notes on how to study from today’s content
Use today’s framework to structure your annotated bibliography and your introduction draft
Prepare to discuss your own research question with peers and identify the most relevant article types (empirical vs review/theoretical)
Practice explaining mediation and moderation with concrete examples from your topic area
Be ready to discuss ethical considerations and real-world implications of your research
Final reminders
Bring three to five topics you want reviewed; we will integrate these into targeted exam prep
Acknowledge the ongoing nature of science; evidence evolves and interpretations may shift with new data
Engage with the podcast material critically; distinguish between association and causation, and be mindful of how media framing can influence public understanding
End-of-class plan
Brief break, then group meetings to plan outside-class work
Opportunity to ask remaining questions and finalize study design planning
Note on math/notation
Key mediation formulas (to memorize):
M = aX + e_M
Y = c'X + bM + e_Y
c = c' + ab
Key moderation model notation (to memorize):
Y = b1 X + b2 Z + b_3 XZ + e
Keep in mind: moderated mediation is optional for this course and considered an advanced topic
Overall takeaway
Today’s session combines practical project planning with a rigorous review of research methods, media literacy, and experimental design
The core skill emphasized: connect empirical evidence to real-world claims while planning rigorous, ethically sound research that can withstand critical examination