Week 2 Analysis Workshop: A Lunchtime Walk and Afternoon Errors

Part 1 — The Big Picture

This week you learned that a correlation tells you how two measures move together, and that only an experiment with random assignment can show a cause. In this workshop you do both halves of that idea yourself. You compute r for a small, made-up set of office volunteers who reported how long they walked at lunch and how many errors they missed on an afternoon proofreading task. You read the number for direction and strength, write the sentence that says "associated with," and name what else could explain it. Then you turn the vague word "focus" into an operational definition, judge it for reliability and validity, and design the experiment that could test whether walking causes fewer errors. Last, you check a chatbot's answers to the same questions and correct what it gets wrong.

Time budget: Plan for about 90 minutes.

Part 2 — The Task

Guiding question: In these eight volunteers, how strongly is walking at lunch associated with afternoon proofreading errors, what is that association allowed to claim, and what study could test whether walking causes fewer errors?

Workshop data: lunchtime walking and afternoon proofreading errors (illustrative)

Source: An illustrative dataset constructed for this workshop, not from a real study: eight hypothetical office volunteers each reported the minutes they walked at lunch and then completed the same afternoon proofreading task, and the errors they missed were counted.

• Rows: 8 volunteers, one row each (the data table in Part 4 also has 8 rows)
• Columns: minutes walked at lunch; proofreading errors missed
• Design: correlational: both variables were measured, and nobody was assigned to walk

Part 3 — Materials & Procedure

Tools and materials

• Google Sheets (free) or Excel
• A document or notebook for your written answers

Ways of working

• compute-and-interpret-r
• operationalize-a-construct
• design-a-correlational-or-experimental-check

Procedure

1. Open a blank sheet. In A1 type Minutes walked and in B1 type Errors. Copy the eight pairs from the data table in Part 4 into A2:B9, one volunteer per row. Tool: Google Sheets (free) or Excel. Check: Rows 2 to 9 hold eight pairs, and the minutes in column A run from 0 to 35 in steps of 5.
2. In D1 type =AVERAGE(A2:A9) and in D2 type =AVERAGE(B2:B9). Copy both means into the results table, then mark each volunteer's row in the data table as above or below the mean walking time. Tool: Google Sheets (free) or Excel. Check: Each mean falls between the smallest and the largest value in its own column.
3. In D3 type =CORREL(A2:A9,B2:B9) and set the cell to show four decimals. Copy r into the results table, then write its direction (positive or negative) and its strength in words. Tool: Google Sheets (free) or Excel. Check: The cell shows a number between −1 and +1 with four decimals.
4. Select A1:B9 and insert a scatter chart. Put minutes walked on the horizontal axis and errors on the vertical axis, and give both axes a title. Describe the pattern in one sentence in words, not by color or position alone. Tool: Google Sheets (free) or Excel. Check: Eight dots appear, one per volunteer, with both axes titled.
5. In your document, write an operational definition of afternoon focus that someone else could repeat, then say whether counting proofreading errors would be reliable and whether it would be valid, with one reason each. Tool: A document or notebook for your written answers. Check: Your definition names an exact, observable procedure with a unit.
6. Complete the design table in Part 4 for an experiment that could test whether walking at lunch causes fewer errors: name the IV, the DV, the control group, how participants are chosen, how they are placed in conditions, and one ethical protection the study must give them. Check: Every row of the design table holds a specific plan, not a single word.

Fallback route

When to use it: Use this route if no spreadsheet is available or the chart will not draw.

Work the same steps by hand on paper. Add each column and divide by 8 to get each mean. For each volunteer, subtract the mean from the minutes and from the errors, multiply the two differences, and square each difference. Add each of the three new columns, then divide the sum of the products by the square root of the product of the two sums of squares to get r, rounding to four decimals. Draw the scatterplot on paper with labelled axes. Complete every scaffold table, answer every analysis question, do the critique step, and submit the same items for equal credit.

This route earns the same credit as the primary route.

Part 4 — The Scaffold

These are the workshop's illustrative data; mark each volunteer against the mean walking time so you can see the pattern before you compute r.

Student scaffold table 1

Volunteer	Minutes walked at lunch	Errors missed	Above or below the mean walking time?
1	0	9
2	5	8
3	10	9
4	15	6
5	20	7
6	25	5
7	30	4
8	35	4

Record your results here so your interpretation rests on numbers you produced yourself.

Student scaffold table 2

Quantity	Your value or words	How you found it
Mean minutes walked
Mean errors missed
r for minutes and errors
Direction of the association
Strength of the association

Plan the experiment that could test whether walking causes fewer errors, one design choice per row.

Student scaffold table 3

Design element	Your plan
Independent variable (IV)
Dependent variable (DV), with its operational definition
Control group
How participants are chosen
How participants are placed in conditions
One ethical protection and why

Part 5 — Analysis Questions

1. Report r for minutes walked and errors missed to four decimals. Say what its sign tells you about direction and what its size tells you about strength.
2. Write the interpretation sentence for these data using the words "associated with." Then name one third variable that could drive both measures, and describe one way the causal arrow could run the other way.
3. Using your design table, explain which feature of your experiment would let it support the claim that walking causes fewer errors, and what random sampling would add that random assignment does not.
4. Give two limits of these data and this measure: one about the sample (eight hypothetical volunteers) and one about whether proofreading errors are a valid measure of focus.
5. Choose one reflection and answer it in 3–4 sentences: (a) a coworker reads your scatterplot and says, "So walking makes people more careful." What would you tell them, and what evidence would change your answer? (b) Think of a habit you believe helps you focus. How would you operationalize focus and design a fair check of your belief?

Part 6 — AI-Critique Moment

Choose one equal-credit critique route. Route A: ask an approved AI tool to critique your completed work, then check its response against your own evidence. Route B is the no-AI equivalent printed below.

Approved tools: Claude, Gemini, ChatGPT, Grok, Microsoft Copilot.

Prompt to paste:

I've pasted my completed Analysis Workshop tables above. In an illustrative dataset, eight volunteers reported minutes walked at lunch and the number of errors they missed on an afternoon proofreading task. The pairs (minutes, errors) are: 0 and 9, 5 and 8, 10 and 9, 15 and 6, 20 and 7, 25 and 5, 30 and 4, 35 and 4. Please do six things. First, compute the correlation coefficient r and interpret it. Second, say whether these data show that walking at lunch reduces afternoon errors. Third, name the independent variable and the dependent variable in these data. Fourth, explain how to choose and place participants in a study that could show whether walking causes fewer errors. Fifth, say whether proofreading errors are a reliable and a valid measure of afternoon focus. Sixth, say whether the statement "walking at lunch improves afternoon focus" is a theory or a hypothesis.

Test the response for these likely failures:

• Chatbots often write that walking at lunch reduces or prevents afternoon errors, turning a correlation into a cause.
• Chatbots often label minutes walked as the independent variable and errors as the dependent variable in these data.
• Chatbots often say that picking volunteers at random from the office is what lets a study show that walking causes fewer errors.
• Chatbots often say proofreading errors must be a valid measure of focus because the count comes out the same each time it is repeated.
• Chatbots often call "walking at lunch improves afternoon focus" a theory.

In 3–5 sentences, say what the chatbot got right, quote at least two sentences it got wrong, and correct each one with evidence from your own tables, your scatterplot, or this week's definitions.

Equivalent route (no AI required)

No AI account is required for this workshop. This route assesses the same objective, Objective 2 (evaluate psychological research methods and ethics, distinguishing correlation from causation, random sampling from random assignment, and identifying sources of bias), takes comparable effort of about the same 90 minutes, uses the same rubric, and earns the same credit. Instead of the chatbot, critique this flawed analysis, printed here in full: "The eight volunteers prove that walking at lunch reduces afternoon errors, because the correlation is strong. Minutes walked is the independent variable and errors are the dependent variable. Because the volunteers were picked at random from the office, the study shows a cause. Counting proofreading errors gives the same number every time, so it must be a valid measure of focus. The idea that walking at lunch improves focus is a well-established theory." Mark it up: quote each flawed sentence, name what is wrong (a correlation read as a cause, variables mislabeled, sampling confused with assignment, reliability confused with validity, or a hypothesis called a theory), cite evidence from your tables or this week's definitions, and write a corrected sentence beneath it. Submit the marked-up analysis in place of the chatbot critique report and share link; it is scored by the same rubric row for the same credit.

Part 7 — What to Submit

• All three completed scaffold tables
• Your scatterplot, or a labelled sketch of it, with a one-sentence description in words
• Your operational definition of afternoon focus with its reliability and validity judgments
• Written answers to all five analysis questions
• The chatbot critique report with its share link, or the equivalent route's marked-up flawed analysis

Deliverable:

A short Analysis Workshop report on walking and afternoon errors: three completed tables (the eight illustrative pairs marked against the mean, your computed means and r with their direction and strength, and your experiment design), a scatterplot or labelled sketch with a one-sentence description, an operational definition of afternoon focus with reliability and validity judgments, written answers to five analysis questions, and a critique of a chatbot's answers to the same questions, or, on the equivalent route, a marked-up correction of the printed flawed analysis. Submit in any accessible file type or as a link; a recorded response may be submitted as audio or video with a transcript, or as text. Choose a written, recorded, or visual form for your answers; every form is graded by the same scoring criteria.

Rubric

The rubric below shows how each submitted part is evaluated; bands are full, partial, or none.

Workshop evaluation criteria

Criterion	Full	Partial	None
Computing and reading r	The means and r are computed from the eight pairs by formula or by hand, r is reported to four decimals, and its direction and strength are stated correctly in words.	r is computed, but a mean is missing or wrong, r is not rounded to four decimals, or the direction or the strength is misread.	No r is reported, or the value is invented rather than computed.
The interpretation sentence and the third variable	The sentence says "associated with" and never claims a cause, names a plausible third variable, and describes a reversed causal arrow.	The sentence avoids a causal claim but omits the third variable or the reversed arrow, or names one only vaguely.	The interpretation claims that walking causes fewer errors, or no interpretation is given.
Operational definition with reliability and validity	Focus is defined as an exact, observable, repeatable procedure with a unit, and reliability and validity are each judged with a specific reason.	A definition is given but is vague or not repeatable, or reliability and validity are judged without reasons or are confused with each other.	No operational definition is given, or reliability and validity are not addressed.
Designing the experimental check	The design table labels the IV and the DV correctly, includes a control group, separates random sampling from random assignment, and names an ethical protection with a reason.	The design is present, but one element is missing, the IV and DV are swapped, or sampling and assignment are blurred.	No design is given, or it cannot support a causal claim and does not say why.
Critique and correction	The chatbot critique report or the marked-up flawed analysis quotes at least two errors and corrects each with evidence; both routes earn this band equally.	The chatbot critique report or the marked-up flawed analysis names an error but corrects it without evidence, or corrects only one; both routes earn this band equally.	No chatbot critique report or marked-up flawed analysis is submitted, or the output is pasted without any check; both routes are scored the same way.

Quality Gate

• Every mean and r in your tables comes from your own formulas or your own hand calculation from the eight pairs, never from memory or a chatbot.
• Your interpretation sentence says "associated with," gives direction and strength, and names a third variable; it never says walking causes fewer errors.
• Your scatterplot or sketch shows eight dots with both axes titled, and its description is in words.
• Your design table labels the IV and the DV correctly and places participants in conditions by random assignment.
• No statistic, study, or source is invented anywhere in the submission.
• The critique quotes the chatbot or the flawed analysis and corrects at least two errors with evidence from your tables or this week's definitions.