1
ATHK1001 ANALYTIC THINKING: ASSIGNMENT 1, 2025

Due date: 11:59pm Friday April 11th (Week 7). Late penalty of 5% per calendar day applies.

Online submission: All submissions are to be made online via the link on the ATHK1001 Canvas
website. All submissions must be a single PDF file. Do not submit files that contain images (except
the graphs can be images), because they may not be readable. Unreadable files may be treated as
non-submissions. Submissions will be checked for plagiarism.

Incorrect submissions: If you discover after the due date but before the closing date that the file you
submitted on was incorrect, then let us know. You may be given the option to resubmit a corrected
version with a 50% penalty or the relevant lateness penalty, whichever is greater.

Word length: 1000 words across all questions (excluding references in Question 13). A penalty of
10% will apply to papers that exceed this limit by up to 10%, a 20% penalty if you exceed up to
20% of the limit, and a 30% penalty if you exceed the limit by more than 20%. There is no leeway
before you are penalized if you exceed 1000 words, and ALL WORDS COUNT. The only exception
are words used in your response to Question 8 (graphs) and Question 13 (reference list). Therefore,
if you include titles or repeat the question text then those words will count. To identify your
answers just give the question number at the beginning of your response to each question.

Total marks: 60 (20% of total grade for class)

Background and Aims

How we best remember things is a question many people are interested in, especially when trying
to study. One common finding is that we have better memory for material we generate ourselves than
material we are simply exposed to. Slamecka and Graf (1978) reported five experiments that firmly
established the generation effect as a memory phenomenon. Their experiments like much of the research
into the generation effect focused on memory for verbal material, so whether generation improves recall
for other types of materials is less examined. So in our class experiment we looked at memory for
numbers and asked whether there is a generation effect for numbers.
Often when the generation effect is tested participants know they are in a memory experiment, and
this may affect what they do. Thus the question arises whether the generation affect is only present when
people expect their memory to be tested. So in our class experiment we tested whether the generation
effect was present both when participants expect a memory test and when they did not.

Method
Participants

Data was recorded from 348 students who attempted the experiment when they participated as
part of a class experiment during tutorials for an analytic thinking course. Consent to have their data
analysed was not given by 101 participants so their data was eliminated from the data set. A further 24
participants were removed because they had not completed the full experiment. This left 223 participants
for analysis. Of the analysed participants 116 identified as female and 104 as male. Only 4 reported being
older than 23 years.

Materials

For the experiment two sets of general knowledge questions with known numerical answers were
constructed: The Generation Set and the Selection Set. The questions for the Generation Set asked about
nine different domains:
What was the total livestock population of [insert animal] in the world in [insert year]
Atomic weight of [insert element]
2
In square kilometers, the area drained by the river [insert river name]
In MILES, length of the river [insert river name]
Total energy consumption per capita (kg of oil equivalent) for [insert country]
External debt per capita ($US) for [insert country]
Expenditure on US TV advertising (US$millons) by [insert company name]
Population of metropolitan area of [insert city]
Daily newspaper circulation of [insert newspaper name]

The questions for the Selection Set asked about a different nine domains:
Is the Total Area (in square kilometers) of [insert country] was [insert number]
In 2013 was the population of [insert country] [insert number]
In 2012 was the total number of homicides in [insert country] [insert number]
Is the length of the [insert river and metric] [insert number]
In 2014 was the Gross National Debt of [insert country and currency] [insert number]
In 2013 were the total number of road deaths in [insert country] [insert number]
In 2015 was the infant mortality rate (per 10000) in [insert country] [insert number]
In 2012 was the yearly consumption of electricity (megawatt hours) in [insert country] [insert
number]
In 2014 were births (per 100000 people) in [insert country] [insert number]

Participants in the experiment completed four tasks: Generation Task, Selection Task, Generation
Recall Task, and Selection Recall Task.
For the Generation Task participants were presented with one question from each domain with the
question completed (i.e., had the “insert” fields filled in) in one of nine different ways. So, a total of nine
different generation questions were asked. Participants recorded their answer by typing a number into a
text box. These were hard questions because the aim was that participants would not generate many
correct answers, and thus the generated answers would be unique to them. If they knew the right answers
then they would not have to recall a new a number that they had not seen before.
For the Selection Task participants were presented with one question from each domain with the
question completed (i.e., had the “insert” fields filled in) in one of nine different ways. So, a total of nine
different selection questions were asked. Participants were presented with either the correct numerical
answer or an incorrect numerical answer which had the same magnitude and first digit as the correct
answer. They indicated whether they thought the presented answer were correct or incorrect by clicking a
button to select “correct” or “incorrect”. About half the presented answers were correct and participants
were told to expect about half the answers to be correct.
The Generation Recall Task showed the participant the same nine questions as they had just seen
for the Generation Task, and they were asked to recall the answer they had given to the same question by
typing the recalled number into a textbox. The Selection Recall Task was the same as the Generation
Recall task except participants were asked to type into the box the number that had been presented as the
answer to the same question when they did the Selection Task.
For all tasks if a participants spent more than 1 minute considering their answer, they saw a pop-
up notice that asked them to respond immediately and that they should not take more than one minute per
question. For all questions participants could not proceed until an answer had been given.

Procedure

During tutorials for the class Analytic Thinking at the University of Sydney participants
completed the experiment individually on computers in a classroom. All participants were randomly
assigned to one of two conditions that determined the order in which the tasks were presented, either
Generation-First or Selection-First.
At the beginning of the experiment participants were told that “In the following task you are
required to give an answer to each question.” They were not told that they would be asked to recall
anything. They then did the experiment in one of two orders:
Generation-First order: Generation task; Generation Recall task; Instruction to expect to be asked
3
to recall answers for next task; Selection task; Selection Recall task.
Selection-First order: Selection task; Selection Recall task; Instruction to expect to be asked to
recall answers for next task; Generation task; Generation Recall task.
So all participants received both the Generation Task and Selection Task and were then asked to
recall the answers they had just generated or seen. In between the tasks participants were told to expect to
recall answers for the next task. So during one task (i.e., the first one they did) participants did not know
their memory would be tested whereas during the other task (i.e., the second task that they did) they knew
their memory would be tested.
After completing the experiment participants indicated whether or not they consented to having
their data included in the data set.

Hypotheses

We proposed five hypotheses to investigate how participants performed in our experiment. We also made
a prediction regarding what result we expected for each experiment.

Hypothesis 1: Mean recall for generated items will by different to mean recall for selected items.
We predicted that generated items will have a higher recall rate than selected items.

Hypothesis 2: There will be a correlation between recall for generated items and recall for selected
items. We predict this correlation will be positive.

Hypothesis 3: For generated items, mean recall will be different depending on whether recall was
for items participants expected to have to recall verse those they did not expect to recall. We
predicted better memory for answers when the participant knew they would have to recall them.

Hypothesis 4: For selected items, mean recall will be different depending on whether recall was for
items participants expected to have to recall verse those they did not expect to recall. We predicted
better memory for answers when the participant knew they would have to recall them.

Results

The data set for this experiment can be found on the Canvas site for ATHK1001 on the page for
Assignment 1 which you can access by clicking on the “Assign. 1” button on our home page. This
assignment description can be found there as well as the Excel file called “Assignment1_dataset.xlsx”.
This Excel file contains all the data necessary for the assignment and has 223 data lines, one for each
participant. Each participant has values for four variables, and the values of each variable are in a single
column of the file.
Column A “id” is an arbitrary identifying number generated by the computer. The “condition”
Column B gives each participant’s condition. “Generation first” means participants were in the
Generation-First order, so they first completed the Generation Task which they did without knowing
they would be asked to recall the answers. “Selection first” means participants were in the Selection-First
order, so they first completed the Selection Task which they did without knowing they would be asked to
recall the answers.
For the Generation Recall Task the proportion of correct items recalled was calculated by
counting the number of times a participant recalled the exact same number as the one they had been given
when they first answered the question, then dividing by the total number of such questions answered.
This proportion is recorded in the Column C of the spreadsheet as “Generation-proportion”. For the
Selection Recall task the proportion of correct numbers recalled was calculated by counting the number
of times a participant recalled the exact same number as the one they had been given as the possible
answer to the question when they were first presented with it, then dividing by the total number of such
questions answered. This proportion is recorded in the Column D of the spreadsheet as “Selection-
proportion”.
4
Note that in order to be marked as having correctly recalled the number correctly ALL the digits
of the recalled number had to match all the digits of the number presented/generated earlier.


WHAT YOU WILL WRITE

Your task is to analyse the data in order to test the four hypotheses proposed above. You will do this by
addressing each of the following 13 questions. Answer all questions with complete sentences, not with
just numbers, notes or tables. You will be penalized if you do not use complete sentences. Do not include
the text of the questions in your assignment (any such text will count towards the word limit), but
you should include the number of the question being addressed.

A note on how many digits to report in your answers. Excel can give answers with a huge string of
digits, many more than are necessary to understand the results, so the question arises of how many digits
to report in your answers? The convention is to use the same number of significant digits (i.e., digits that
are not zero in the numbers) as the data used in the calculation, and all the data has three significant
digits. So you should report all statistics to three significant digits.

1) For both Generation-proportion and Selection-proportion report their means and standard deviations.
(4 marks)

2) Based on the means you calculated in Question 1 use a t-test to test Hypothesis 1, that mean recall for
generated items will by different to mean recall for selected items. Report the p-value for the t-test and
state clearly the conclusion regarding Hypothesis 1, and whether our prediction was supported. (Note
that you will practicing using Excel to test hypotheses in tutorials in Week 5.) (3 marks)

3) Calculate the correlation between Generation-proportion and Selection-proportion and test Hypothesis
2. Report the p-value for the t-test and state clearly the conclusion regarding Hypothesis 2, and
whether our prediction was supported. What does this tell us about the relationship between these two
measures of memory? (6 marks)

4) Calculate the Generation-proportion means and standard deviations for the Generation-first order and
for Selection-first order conditions. (4 marks)

5) Based on the means you calculated in Question 4 use a t-test to test Hypothesis 3, that for generated
items mean recall will be different depending on whether recall was for items participants expected to
have to recall or for those they did not. Report the p-value for the t-test and state clearly the
conclusion regarding Hypothesis 3, and whether our prediction was supported. (3 marks)

6) Calculate the Selection-proportion means and standard deviations for the Generation-first order and for
the Selection-first order conditions. (4 marks)

7) Based on the means you calculated in Question 6 use a t-test to test Hypothesis 4, that for selection
items, mean recall will be different depending on whether recall was for items participants expected to
have to recall or for those they did not. Report the p-value for the t-test and state clearly the
conclusion regarding Hypothesis 4, and whether our prediction was supported. (3 marks)

8) Present two graphs: a histogram for Generation-proportion, and a histogram for Selection-proportion.
These graphs display the distribution of these two variable. (6 marks)

9) Based on the graphs you produced in Question 8, to what extent do these two variables have a normal
distribution? Do these distributions indicate any problems regarding your tests of Hypotheses 1 and 2?
(4 marks)

5
10) Identify three different issues with the way we collected data which could limit our ability to draw
conclusions from it. These issues should relate to one or more of the hypotheses. Clearly differentiate
the three issues as “Issue 1”, “Issue 2” and “Issue 3” and explain how each of these issues relates to a
data collection consideration raised in ATHK1001 lectures, readings or tutorials. For each issue
suggest a way it might be resolved in future research on this topic or if it cannot be resolved then
explain why. To receive full marks you must specify how and why this issue applied to this
experiment. Broad issues, such as the possibility of using a larger sample which could be said of
almost any experiment, will not receive full marks. (12 marks)

11) Summarize what you think the data analysis you have carried out for this assignment tells us about
the generation effect. In doing so you should cite Slamecka and Graf (1978). Explain your responses
with explicit reference to the results of your testing of the hypotheses and possibly the issues you
raised in Question 10. Be specific to this experiment. (5 marks)

12) Summarize what you think the data analysis you have carried out for this assignment tells us about
the effect of knowing that your memory will tested. Explain your responses with explicit reference to
the results of your testing of Hypotheses 3 and 4 and possibly the issues you raised in Question 10. Be
specific to this experiment. (4 marks)

13) Include a reference section which lists the full reference for any paper you have cited when
addressing these questions. You must include Slamecka and Graf (1978) in this question, and include
other references when appropriate. You should use APA style for citations and references, but we will
accept other standard journal article referencing formats. (2 marks)


Reference

Title: The generation effect: Delineation of a phenomenon.
Author: N. J., Slamecka, and P. Graf
Source: Journal of Experimental Psychology: Human Learning and Memory. 1978. Volume 4. Pages
592–604.

THESE ARE NOT IN A STANDARD REFERENCING FORMAT; YOU WILL NEED TO
REFORMAT THIS FOR QUESTION 13 USING A STANDARD FORMAT (E,G., APA STYLE)

Note that these papers provide background so you do not have to understand every aspect of them. Also,
you do not have to use other sources for answering these questions, but if you do then you must correctly
cite and reference these sources.

Formatting Recommendations

Our preferences
 Use the font “Times New Roman”, 12-point size, and double-space all the lines.
 Indent the beginning of each paragraph using one tab space.
 Use APA referencing style



学霸联盟