PSYC122 - Weak 16 to 19

?
• Created by: Lucylight
• Created on: 13-03-23 13:35
Week 16
Hypothesis, measurement and associations
1 of 23
Aims throughout weeks 16-19
1) To connect the ideas of the scientific method, measurement and hypothesis testing and modern reproducible open science together.
2) To deepen and broaden your skills working with correlations and linear models.
3) Develop your critical thinking skills
2 of 23
What is the project?
We're applying these skills to a live investigation about understanding of medical information. People can estimate they understand more than they do and assuring people understand their diagnosis or how to stay in good health is critical.
3 of 23
Learning targets for this week include:
Concepts – associations: correlations, estimates and hypothesis tests
Skills – visualizing variation and covariation
Skills – writing the code
Skills – estimating correlations
Skills – interpreting and reporting correlations
4 of 23
Explain the derivation chain and what it applies to in good statistical science?
We need to think causally about predictions and measurements, and instead of simply looking if a statistic is significant or not, we need to start with more solid and better theory in which we can build clear testable predictions from direct assumptions.
5 of 23
The Derivation Chain
1) Concept formation Causal model
measurement Auxiliary assumptions
Both go into statistical predictions
Testing hypothesis
6 of 23
Explain the first step of the derivation chain (Meehl, 1990; Scheel et al., 2021)
1) The first step about considering an hypothesis productively is to develop your theory. This involves establishing, understanding and developing your concepts and the (literary supported- usually/best) assumptions about causality.
7 of 23
Explain the second step of the derivation chain (Meehl, 1990; Scheel et al., 2021)
2) Clearly identify how you psychological concepts will be measured.
8 of 23
Explain the third step of the derivation chain (Meehl, 1990; Scheel et al., 2021)
3) Identify you assumptions about how you will get from the theoretical concepts to observable data i.e. how you will operationalise/make variables to test the concept.
9 of 23
Explain the fourth step of the derivation chain (Meehl, 1990; Scheel et al., 2021)
Specify theoretical predictions.
10 of 23
Explain the fifth and final step of the derivation chain (Meehl, 1990; Scheel et al., 2021)
Link you theoretical predictions to specific statistical tests that may support or contradict them.
11 of 23
What is validity?
Validity is the accuracy of a measure - particularly referring to a populations true effect. True variation i.e. reflective variation.
12 of 23
What makes a test/measurement valid?
A test is valid for measuring an attribute if the attribute a) exists and b) variations within it attribute causally/ produce variations in the outcomes of the measurement procedure.
13 of 23
What are the research questions linked with our live health comprehension project?
1) What person attributes predict success in understanding?
2) Can people accurately evaluate whether they correctly understand written health information?
14 of 23
What is the research question with IV and DV?
Is there an association and significant variation between comprehension and vocabulary?
15 of 23
How are we measuring the DV?
Questionnaires/survey's measuring there reading comprehension with text and answer and vocabulary Knowledge (Shipley) and health literacy (HLVA) for experience.
16 of 23
Identify the theoretical context?
Research has demonstrated an individuals understanding of text varies due to language experience and reasoning capacity.
17 of 23
Can an individual accurately evaluate their own understanding using self-reporting measures?
Subjectivity of judgements - is their understanding though so we do want to see what they think vs what they do.. however,,
18 of 23
What is the outcome, predictor and the linear model?
The outcome is the dependent variable that is assumed to vary in response to the predictor, independent variable. The linear model is a regression analysis which quantifies the association between variables and how accurate they are.
19 of 23
Measurement Methods - discuss the importance of covariance (correlation measure) with different scales, histograms and scatterplots.
Covariance is the measure of how much factor y varies with x, using different measuring scales i.e. like SHIPLEY and HLVA then using them or any other pairs of factors together dividing covariance by standard deviation produces the correlation without var
20 of 23
R code for correlation test is?
cor.test(clearly.one.subjects\$mean.acc,
clearly.one.subjects\$mean.self,
method = "pearson")
21 of 23
What to read for correlation analysis?
Direction and strength - cor
P-value to indicate if it's a significant correlation.
(r(167-df) =.49 , p <.001
22 of 23
What does correlation analysis do?
Reports the strength and direction of the association.
23 of 23

Other cards in this set

Card 2

Front

Aims throughout weeks 16-19

Back

1) To connect the ideas of the scientific method, measurement and hypothesis testing and modern reproducible open science together.
2) To deepen and broaden your skills working with correlations and linear models.
3) Develop your critical thinking skills

Card 3

Front

What is the project?

Card 4

Front

Learning targets for this week include:

Card 5

Front

Explain the derivation chain and what it applies to in good statistical science?