Mini-Dissertation Submission

Your Mini-Dissertation final submission must comprise ALL of the following COMPULSORY elements:

Recap of 2x2

3 ‘flavours’ of 2x2 ANOVA

2bx2b - Between-subjects/Factorial ANOVA

2wx2w - Within-subject/Repeated Measures ANOVA

2bx2w or 2wx2b - Mixed ANOVA

Formalise your individual design

More specific

How you blend 2x2 designs as a group

Let’s say your group was interested in Academic Self-Handicapping and features of personality and academic background contributing to this phenomenon.

It makes sense that the bulk of the efforts of your group go in to measuring Academic Self-Handicapping.

Individual 2x2

The effect of Conscientiousness and Self-Efficacy on Academic Self-Handicapping

Rest of the group:

The effect of IV(A) and IV(B) on Academic Self-Handicapping 

Overall data collection

• IV(A) Conscientiousness Questionnaire
• IV(B) Self-Efficacy Measure
• (DV) Academic Self-Handicapping Test
• (+) Procrastination as IV(B) (Person 2 complete)
• (~) Person 3 Procrastination - IV(A)/Self-Efficacy - IV(B) & ASH (Person 3 done)

If you develop a more elaborate ‘task’

such as a memory test, with easy and difficult stimuli sets

• This is a 2-level IV already (let’s call it IV(B)
• Group members could either
  • Swap in different personality measures or attributes as IV(B)
  • OR look at different DVs (Accuracy, Error, Confidence, Speed, reported difficulty)
  • OR you could overlay another IV - to introduce practice (but this may slightly separate datasets meaning more data would be required (could be conceived as a poorer or better planned design!)

Do not expect yourself to be able to solve this now

• Once we come back after reading week, and we start building, you will know more about your chosen topic.
• You will have an idea of preferred methods
• Your PT and LT will be able to refine and discuss ideas
• The Build and Ethics Submission are both points which we can be utterly certain your design works and is achievable
• This is why you need to talk, keep notes, and should be strategic with the Critical Proposal

‘Operationalisation’ of variables

Operationalisation of variables requires a consideration of the reliability and validity of the method of operationalisation.

Operationalisation of variables also requires specification of the scale of measurement: nominal, ordinal, interval, or ratio.

Finally, operationalisation of variables can also specify details of the measurement procedure.

See Howitt and Cramer Chapter 3.3 (Box Research Example - Condon & Crano 1988)

Attitude similarity and interpersonal attraction

Their DV (consider pros and cons)

This person would like (dislike) me

This person would like (dislike) working with me in an experiment

The Student Room (question)

The Student Room (answer 1)

The Student Room (answer 2)

Of great importance that you reflect on this for your MD too

How have you defined your variables of interest?

• IVs and DVs

How have you measured or categorised your variables of interest?

• e.g. Social Media use

• Gender

• Frequency or low/high extraversion

The research process

Quantitative Research

A systematic examination of relationships between variables

Types of Variable

Independent Variable

• Experimental - The variable I manipulate

• Non-Experimental / Quasi-Experimental - Comparison between pre-existing groups

Dependent Variable

• The variable you measure, that you propose to be influenced by a manpulation of the IVs
• The [D]ata{.shout}

Types of Measurement

Nominal/Categorical

• Male/Female/…

• Vegan / Vegetarian

• Smoker/Non-Smoker

Types of Measurement

Ordinal

Numbers representing a rank position in a group

Not representative of an actual definite number/score/value - without information about the ‘gap’ between numbers

• First, second, third

• Tallest/Shortest

Types of Measurement

Interval

Numbers represent equal units giving information about the ‘gap’ between numbers

• Temperature

• Psychological Scales

Types of Measurement

Ratio

Interval measurements with an absolute zero, of equal units,

• Weight

• Length

• Time/Reaction time*

Median Splits

‘Cutting’ a distribution in half at the mid-point - with 50% on each side of the cut

Median Splits can be considered problematic

It is a key learning outcome that you are able to perform a standardised analysis, specifically, the 2x2 ANOVA with any necessary assumption checks and post-hocs + plots

Think about this

Operationalisation, measurement and definitions impact…

Summary

You should think carefully about:

• How you define your variables - this is probably a part of the introduction that students DON’T think about enough

• How you measure or categorise your variables (IVs and DVs) - this is probably the single thing I look at first when peer-reviewing research!

• How well your manipulation does what it claims to. Does you manipulation bring the thing it proposed to to life well?

Comparing Experiment Modalities

• Online Surveys and Experiments
  • Qualtrics - For surveys, stimulus presentation that don’t require accurate timing
  • Gorilla.sc - For ‘experimental’ tasks with complex stimulus > presentation or that require response time measures or push-button > responses etc.
• Face-to-Face Experiments

Online Surveys and Experiments with Qualtrics

Overview

• A powerful tool for creating and distributing online surveys
• Widely used in psychological research
• We have an institutional licence and so it is FREE
• No specialist support - has excellent online support

Pros of Qualtrics (or online surveys)

Accessibility

• Reach a wide, diverse audience (if necessary)
• Accessible by anyone with a phone or laptop etc

Cost-Effective

• Low ‘cost’ compared to face-to-face

Data Management

• Automated data collection, scoring and storage
• Relatively easy to analyze and export data - but not always simple

Customization

• Highly customizable surveys and experiments (video, photos, audio, vignettes)
• Supports various question types and logic

Cons of Qualtrics

Limited Human Interaction

• Lack of personal interaction can affect responses
• You put a huge amount of trust in the participant
• Not always easy to manage ethical challenges properly

Technical Issues

• Dependence on internet connectivity and user tech skills
• An error in the build and you are in trouble ####

Cons of Qualtrics (continued)

Response Authenticity

• Higher risk of dishonest or inattentive responses (maybe?)
  

Sampling Bias

• Potential for non-representative samples

Introduction to experiments using Gorilla.sc

Gorilla.sc: highly controlled online Psychological Research

• A comprehensive online tool for behavioral research
• Facilitates creation, deployment, and analysis of experiments

Core Features of Gorilla.sc

User-Friendly Interface

• Intuitive design for researchers of all levels
• Drag-and-drop experiment builder

Requires ‘tokens’ to run

• You will need to apply for ‘tokens’ to apply for ethics, and to run the study.
• There is usually a limited supply as there is a cost associated with this.

Versatile Experiment Design

• Supports a wide range of experimental paradigms
• Customizable to suit various research needs

Experiment Creation with Gorilla.sc

Building Blocks

• Use ‘widgets’ and ‘tasks’ to construct experiments
• Easily integrate surveys, quizzes, and cognitive tasks

Advanced Customization

• Incorporate complex experimental logic
• Customize with CSS and JavaScript for unique requirements

Data Collection and Analysis

Real-Time Data Collection

• Gather data securely and efficiently
• Access participant responses in real-time

Analytical Tools

• Built-in tools for basic data analysis
• Export data for advanced analysis in other software e.g. SPSS

Collaboration and Sharing

Team Collaboration

• Share experiments and data with team members
• Collaborate on experiment design and analysis

Participant Recruitment

• Still needs to be managed via sources outside of Gorilla

Gorilla.sc in Psychological Research

Broad Application

• Suitable for cognitive, social, and clinical psychology
• Ideal for remote and large-scale studies

Impact on Mini-Dissertations

• Increases the reach of psychological studies requiring complex stimulus or timing
• Not easy to get up and running with this tool
• No specialist support - has excellent templates and online support

Face-to-Face Experiments

Overview

• Traditional method of conducting experiments
• Involves direct interaction with participants

Pros of Face-to-Face Experiments

Enhanced Interaction

• Direct human interaction enriches data quality

Controlled Environment

• Better control over experimental conditions

Immediate Feedback

• Opportunity for immediate clarification and feedback

Participant Authenticity

• Lower risk of false responses (maybe?)

Cons of Face-to-Face Experiments

Higher Costs

• Greater resource and time investment (maybe?)

Limited Reach

• Restricted to participants’ geographical location

Time-Consuming

• Scheduling and conducting sessions takes time

Potential Biases

• Risk of experimenter or social desirability biases

Conclusion

Choosing the Right Method

• Depends on research goals, resources, and target population
• Both methods offer unique advantages and challenges

Tailoring to Research Needs

• Consider the nature of your study and participant accessibility
• Balance between quality, ‘cost’, and reach

Critical Proposal Target Paper ‘approval’

Ask you Lab Tutor for ‘thumbs up/down’

Prepare for Personal Tutor Meeting

Get ready to present your initial group membership, area of interest and CP papers to your Personal Tutor

If you PT comes before the lab… Better think about that ASAP!

See if you can start identifying candidate methods

Talk amongst your group and discuss the merits of particular methods and discuss this with your LT and PT