CH 2 - Reliability, Validity, and LOTS

Question 1

What are the 3 main types of research design in personality?

Answer 1

1. Case Studies
2. Correlational Studies
3. Experiments

Question 2

What is a case study in personality research?

Answer 2

Systematic analysis – of a single person – or a – small group.

Question 3

Why can’t case studies always reveal universal principles?
– Give example.

Answer 3

Because – 1 person is not representative – of – the general population.

• Example = Findings – from 1 female subject – don’t necessarily apply– to all women.

Question 4

How do case studies relate to McAdams’ Level III (Identity)?

Answer 4

• Allow deep insight – into a person’s life story– and – identity development.
• Capture – past experiences, – values, – and personal narratives.

Question 5

What are common examples of case study topics?

Answer 5

• Brain damage (e.g., patient H.M.).
• Clinical cases (e.g., unique psychopathologies).
• Rare conditions or extraordinary abilities.

Question 6

Why were case studies dominant in early psychology (during Frued’s time)?

Answer 6

• Statistics – and – experimental methods – were not – yet developed.
• Offered the only systematic way – to study – individual minds – in detail.

Question 7

What are the Strengths of Case Study?

Answer 7

1. Relatively inexpensive.
2. Observations provide a good starting point (I hear you say something, and other six people say the exact same thing; hence, observation is a good starting point).
3. Provides an – in-depth – look at personality dynamics – in one person.

Question 8

What are the Limitations of Case Study?

Answer 8

1. One person – is NOT representative – of the entire population.
2. Difficult to – run formal statistics – on the data (case studies).
3. Time-consuming.

Question 9

What are the Strengths & Limitations of Case Study?

Answer 9

STRENGTHS:
1. Relatively inexpensive.
1. Observations provide a good starting point (I hear you say something, and other six people say the exact same thing; hence, observation is a good starting point).
1. Provides an – in-depth – look at personality dynamics – in one person.

LIMITATIONS:
1. One person – is NOT representative – of the entire population.
1. Difficult to – run formal statistics – on the data (case studies).
1. Time-consuming.

Question 10

What are Correaltion Studies?

Answer 10

The studies that – examine how – 2 variables relate – to one another.

Question 11

What is a correlation?

Answer 11

A measure – of the extent – to which two variables – change together.

Question 12

What is Correlation represented by?

Answer 12

Correlation Coefficient (r)

Question 13

What is Statistical definition of Correlation Coefficient?

Answer 13

Statistically, – this is – a ratio of covariability – to (divided by) total variability.

• Correlation coefficient = ratio of covariability / total variability

Question 14

What is English definition of Correlation Coefficient?

Answer 14

This value represents – how much – of the total measured change – in both variables – is due to their relationship.

Question 15

What does the ABSOLUTE VALUE of the correlation coefficient indicate?

–EXAM Ques

Answer 15

The larger – the absolute value of this number is, – the STRONGER the relationship – between the 2 variables.

Question 16

What is Correlation Coffeciant represented by?

Answer 16

r

Question 17

What must r always be between?

Answer 17

–1.00 and +1.00

Question 18

Why must r always be between –1.00 and +1.00?

Answer 18

Because – the numerator (covariability) – CANNOT exceed – the denominator (total variability). —- Any ratio – greater than 1 – or less than –1 – is mathematically impossible.

Question 19

See picture.

Answer 19

Question 20

What does the + and – sign infront of r indicate?

Answer 20

The direction of the relationship.

• + (positive): As one variable goes up, – the other goes up.
• – (negative): As one variable goes up, – the other goes down.

Question 21

What does the absolute value of r tell us?

Answer 21

The strength of the relationship:

|r| close to 1.0 = strong relationship. E.g., .9
|r| close to 0.0 = weak relationship. E.g., .05

Question 22

How do you interpret no to moderate correlation strengths?

? = No relationship

? = Small relationship

? = Moderately small relationship

? = Moderate relationship

Answer 22

r = 0: No relationship.

r < 0.10: Small relationship.

r = 0.10–0.20: Moderately Small relationship.

r = 0.20–0.30: Moderate relationship.

Question 23

How do you interpret moderately strong to strong correlations?

? = Moderately Strong relationship
? = Strong relationship

Answer 23

r = 0.30–0.40: Moderately Strong relationship.

r = 0.40–0.50: Strong relationship.

Question 24

What about very strong or extremely strong correlations?

? = Very strong relationship (RARE).

? = Extremely RARE.

Answer 24

r = 0.50–0.60: Very strong relationship (RARE).

r > 0.60: Extremely RARE.

Question 25

What does the following suggests? ***r*** > **0.60**: **Extremely RARE**.

Answer 25

Often suggests -- the **two** “**DISTINCT**” **variables** -- may **actually** be **measuring** -- the **same construct**.

Question 26

How are correlations best visualized?

Answer 26

Using **Scatterplots**

Question 27

What is a Scatterplot?

Answer 27

A **graph** -- in which -- **one variable** is on the **X-axis** (**horizontal axis = IV)** -- and -- the **other** is on the **y-axis** (**vertical** **axis = DV**).

Question 28

What does Each point on the graph represents?

Answer 28

A **person** -- **in** the **study**.

Question 29

What are **strengths** of correlational studies?

Answer 29

1. **Investigate** -- **multiple variables** -- at **once**. 1. **No need** -- for **variable manipulation** * ***Ethical study*** of ***non-manipulable factors*** (e.g., ***(study depression and social media use = here I can NOT manipulate depression, i.e., tell people X is depressed and Y is not***). 1. Examine **real‑world relationships** and can extend to **predictive analyses** (***Can extend correlation study to predictive study. Where I can predict depression based on social media use.***)

Question 30

Why are correlational studies useful for ethical research? -- Give example.

Answer 30

They **let** us **study** **variables** we **cannot manipulate**. * Example: study depression and social media use = here I can N**OT manipulate depression**, i.e., **tell people X is depressed and Y is not**.

Question 31

How can correlational studies lead to predictions?

Answer 31

By **identifying relationships**, -- you can **use one variable** (**IV**) to predict **another** (**DV**) -- in future samples (e.g., ***predict depression*** based on ***social media use***).

Question 32

What is an illusory correlation?

Answer 32

The **tendency** -- for **people** -- to **perceive** a **pattern** -- that is **not there**, -- or **overestimate** its **strength**.

Question 33

What causes illusory correlation?

Answer 33

1. **Selective attention**. 1. **Regression** **toward** -- the **mean**.

Question 34

What is **Regression** **toward** -- the **mean**?

Answer 34

The **tendency** -- for **extreme scores** or **events** -- to **fall back** (**regress**) --**towards** the **average**.

Question 35

Give example -- **Regression** **toward** -- the **mean**.

Answer 35

Question 36

Why does “correlation ≠ causation”?

Answer 36

Because --a **relationship** -- between **2 variables** -- **doesn’t prove** -- that **one causes** the **other** —— there might be **third** **variables** causing that **relationship**.

Question 37

What are the Limitations of Correlation Studies?

Answer 37

1. **Illusory correlations**. 1. **Correlation** is **NOT** **causation**!!! **Correlation ≠ Causation**.

Question 38

What does a correlation of +0.45 between happiness and number of friends mean?

Answer 38

There is -- a **strong positive relationship**: -- as ***one goes up***, -- the ***other tends to go up***.

Question 39

Interpret the following correlation: * The correlation between happiness and how many friends you have is +0.45

Answer 39

**3 possibilities**: 1. **Being happier** -- causes -- you to **have more friends**. 1. **Having more friends** -- causes -- **you to be happier**. 1. A **strong correlation** -- with some **third variable** -- is causing -- the **relationship between** -- these **two** * Called the ***Third variable problem*** (***Extroversion*** might make people both ***happier*** and ***more sociable).***

Question 40

What problem does the Third variable problem produces?

Answer 40

A **statistic** **phenomenon** -- called **Spurious Correlations**.

Question 41

What is a spurious correlation?

Answer 41

**Refers** -- to a **relationship** -- between **two** **variables** due to: * The **presence** of a **third variable**. * **Chance**.

Question 42

How does the third variable problem create spurious correlations?

Answer 42

A hidden factor -- causes both variables -- to change together, -- making it appear -- as if they’re directly related -- when they’re not.

Question 43

Give 3 fun examples of a spurious correlation.

Answer 43

Question 44

What is an Experiment?

Answer 44

A **type** of **data collection** -- in which **1** or **more variables** (***IV***) -- are **manipulated** -- to **observe** -- the **effects** on **another variable** (***DV***)

Question 45

Experiment is the **ONLY** type of data collection which can make which what?

Answer 45

**Cause** and **Effect** **Claims**.

Question 46

What is an Experimental Group?

Answer 46

The **group** -- **exposed** to the “**treatment**” or **manipulation** -- of the **IV**.

Question 47

What is a Control Group?

Answer 47

The **group** -- **NOT** **exposed** to the **manipulation**, -- which will **serve** for **comparison** **purposes**.

Question 48

Why are participants randomly assigned to groups in experiments?

Answer 48

To **control** -- for **external** **variables** -- and -- **ensure** both **groups** are **similar** -- before **treatment**.

Question 49

What is Counterbalancing and why is it used?

Answer 49

To **control** -- for **order** **effects**, like ***fatigue*** or ***practice***.

Question 50

What are the Strengths of Experiments?

Answer 50

1. Can **provide** -- **cause** -- and -- **effect claims**. 1. Multiple **IVs** -- can be **investigated** at **once**. 1. **Quasi-experiments** (compare ***Extrovert*** vs. ***Introverts***)

Question 51

What are the Limitations of Experiments?

Answer 51

1. Can be -- **VERY** **expensive**. 1. **Some variables** -- **can’t** be **manipulated** 1. It is **impossible** -- to **control** for -- **all** **confounding variables**

Question 52

What are the Strengths and Limitations of Experiments?

Answer 52

STRENGTHS: 1. Can **provide** -- **cause** -- and -- **effect claims**. 1. Multiple **IVs** -- can be **investigated** at **once**. 1. **Quasi-experiments** (compare ***Extrovert*** vs. ***Introverts***). LIMITATIONS: 1. Can be -- **VERY** **expensive**. 1. **Some variables** -- **can’t** be **manipulated** 1. It is **impossible** -- to **control** for -- **all** **confounding variables**

Question 53

What is a Quasi-experiment?

Answer 53

A study -- that compares -- pre-existing groups, -- like extroverts vs. introverts, -- where random assignment -- is not possible.

Question 54

What does Null Hypothesis Significance Testing (**NHST**) help determine?

Answer 54

To **determine** -- whether -- results are **due** to **chance** -- or -- **actual differences**.

Question 55

Example: * I want to know whether my new drug improves exam scores. I will give **one group my new drug (experimental group)** and ***one group a sugar pill (control group)***

Answer 55

Question 56

What is the Null Hypothesis (H₀)? -- Give example.

Answer 56

**Refers** -- to the **hypothesis** -- that there **is NO difference** -- **between your groups**. * Example = ***Both my groups will have the same score on the exam***

Question 57

What is the Alternative Hypothesis (H₁)? -- Give example.

Answer 57

**Refers** -- to the **hypothesis** -- that **there IS** -- a **difference** -- **between your groups**. * Example = ***The two groups will not have the same score on the exam***.

Question 58

What is a one-tailed hypothesis?

Answer 58

Where we -- **predict** -- a **specific** **direction** -- of **effect**.

Question 59

When do we use a one-tailed hypothesis?

Answer 59

Typically used -- when **replicating** a **study** -- or -- **testing** a **very specific prediction.**

Question 60

What is a two-tailed hypothesis?

Answer 60

Where -- we **predict** a **difference**, -- but **DON'T** know -- in **which direction** -- the **effect will be**.

Question 61

Why is it called Null Hypothesis Significance Testing?

Answer 61

Because we are -- **testing** -- whether -- we **can support** -- or -- **reject** the **null hypothesis**.

Question 62

Do researchers manually set the alpha level for their test?

Answer 62

**Yes**

Question 63

What is the alpha level (α) in hypothesis testing?

Answer 63

A **probability** **threshold** -- which **represents** -- the **probability** -- that -- **your observed results** -- will be **due** to **chance**.

Question 64

What does an alpha level of 0.05 mean?

Answer 64

* There is -- a **5%** **chance** -- that -- the **observed results** -- occurred -- due to **random chance**. * **Results** with a **p-value** below **0.05** -- are **considered** -- **statistically significant**.

Question 65

see picture.

Answer 65

Question 66

Do we REJECT or ACCEPT the null hypothesis, when results are statistically significant?

Answer 66

We **REJECT** the **null hypothesis** (**H₀**).

Question 67

What do we claim when the results are statistically significant? -- Give example.

Answer 67

We can claim -- **there IS a true difference -- between groups** -- (***not just due to chance***). * Example: A ***drug did significantly impact exam scores.***

Question 68

see picture.

Answer 68

Question 69

Does statistical significance mean the results weren’t due to chance?

Answer 69

**No** — it only **means** there’s a **low probability** (e.g., ***less than 5%***) that the **results** were **due** to **chance**. ---- There’s still a **risk they were**.

Question 70

What is a Type 1 error?

Answer 70

**Rejecting** -- the **null hypothesis** -- when it is **actually true**. * ( This means the study concludes **there’s an effect**, -- but **there isn’t one**.)

Question 71

What does alpha (α) represent in terms of error?

Answer 71

Alpha is -- the **exact probability** -- of -- **making** a **Type 1 error**. * Example: **α** = **0.05**→ **5% chance** of **rejecting** a **true null hypothesis** (**H₀**).

Question 72

What is a Type 2 error?

Answer 72

**Failing** to **reject** the **null hypothesis** when it is **actually** **false**. * (saying **there's no effect** -- or -- **difference** when **one actually exists**).

Question 73

Answer 73

Question 74

What is statistical significance?

Answer 74

* A **measure** -- of **how likely** it is -- that **results** -- occurred by **chance**. * **Statistically significant** = **very low probability**-- the **result** -- was **random**/**due to chance**.

Question 75

What alpha level is commonly used to define statistical significance?

Answer 75

**0.05** (**5%**)

Question 76

What is practical significance?

Answer 76

* The **real-world impact** -- or -- **importance** of your **findings**. * **Even statistically significant** results -- may have **minimal practical value**.

Question 77

Can a very large sample size produce statistically significant results even if the effect is small?

Answer 77

**Yes**. * **Large samples** -- can **detect** -- **tiny effects** -- that may **not** be -- **practically meaningful**.

Question 78

Does statistical significance imply practical significance?

Answer 78

**No**

Question 79

Give an example of practical significance being low despite statistical significance?

Answer 79

**Facebook study**: People shown **positive** vs. **negative** **news** -- had **posting behavior** **changes less than 1%** (i.e., people who saw +ve news posted +ve post and people who saw -ve news posted -ve post were less than 1%) — **statistically** **significant**, but **practically trivial**.

Question 80

Answer 80

Question 81

Reliability -- from here on--

Answer 81

-- from here on--

Question 82

What is Reliability?

Answer 82

The **degree** --,**to which** -- your **test** -- produces -- **consistent results**

Question 83

Give 2 examples of high reliability.

Answer 83

* **Professor X** -- starts class at **10AM** -- **every week** —— shows **consistent timing**, -- hence **reliable**. * An **IQ** **test** -- is **reliable** -- if the **same person** -- gets the **same score** -- on the **test** -- **two weeks** -- in a **row**.

Question 84

Example of low reliability/unreliable.

Answer 84

**Professor Y** -- starts class **sometime** between **10:00 AM** and **10:20 AM** -- **each week** —— **inconsistent**, -- hence **unreliable**.

Question 85

Is reliability a type of correlation?

Answer 85

**Yes**. * We want the **values** -- to be -- **as** **HIGH** **as possible**

Question 86

What is a true score?

Answer 86

A **person’s** -- **actual** **score** -- on a **test**.

Question 87

Can we ever know a person’s true score?

Answer 87

**No**. * A **true score** -- is **unobservable** -- and can **never** be -- **directly measured**.

Question 88

What is a measured score?

Answer 88

The **score** -- which the **person** -- **actually obtains** -- on the **test**

Question 89

What is measurement error?

Answer 89

The **difference** between -- the (***unobservable***) **true score** -- and -- the **measured score**.

Question 90

What do reliable tests have in common?

Answer 90

They have -- **Low measurement error**, -- meaning the **measured score** is **close** -- to -- the **true score**.

Question 91

see picture

Answer 91

Question 92

Why is having reliable measures important?

Answer 92

Because we **can’t test the same person 50 times** .

Question 93

What happens if a measure is **unreliable**?

Answer 93

Even if **someone's** **true score** is **100**, -- the **measured scores** will **vary** **widely**, -- **making** the **results** **inconsistent**.

Question 94

How can we simulate repeated testing without retesting?

Answer 94

By **increasing** -- the **number of items** -- that **measure** -- the **same thing** on a **test**.

Question 95

What is the **Principle of Aggregation**?

Answer 95

**Combining multiple observations** -- (e.g., multiple items/questions) -- **increases** **reliability** -- of the **measurement**.

Question 96

Reliability is a Special Type of Correlation -- see picture--

Answer 96

Question 97

In **reliability**, we are **assessing** the **SAME** **variable**!

Answer 97

--

Question 98

What r-value indicates **excellent reliability**?

Answer 98

***r*** > **0.90** = **Excellent reliability** ✅

Question 99

What r-value indicates **very good reliability**?

Answer 99

***r*** between **0.85** and **0.90** = **Very good reliability**👍

Question 100

What r-value indicates **good reliability**?

Answer 100

***r*** between **0.80** and **0.85** = **Good reliability** (***0.80 is the general cutoff***) ✔️

Question 101

What r-value indicates **okay reliability**?

Answer 101

***r*** between **0.70** and **0.80** = **Okay reliability** 😐

Question 102

What r-value indicates **bad reliability**?

Answer 102

**r** between **0.60** and **0.70** = **Bad reliability** ❌

Question 103

What r-value indicates **very bad reliability**?

Answer 103

***r*** < **0.60** = **Very bad reliability** — ***rethink your measure***! 🚨

Question 104

What are the 4 types of Reliability?

Answer 104

1. **Test-retest** (across ***time***) 1. **Parallel Forms** (across ***tests***) 1. **Split-half** (across ***items***) 1. **Inter-judge** (across ***observers***)

Question 105

What is Test-Retest Reliability aka?

Answer 105

**Reliability across time**

Question 106

What is Test-Retest Reliability?

Answer 106

The **reliability** -- of your **measure** -- **given** to the **same** -- **person** -- at **two different time points**.

Question 107

Give an example of Test-Retest Reliability.

Answer 107

Question 108

What is Parallel Forms Reliability aka?

Answer 108

**Reliability Across Tests**

Question 109

What is Parallel Forms Reliability?

Answer 109

The **reliability** -- of -- **two different** **versions** -- of a **test** -- **given** to the **same** **person**.

Question 110

Give 2 examples of Parallel Forms Reliability.

Answer 110

* **Form A** and **Form B** of **tests** -- given to the **same person** named **Sam**. * A professor gives -- **Exam 1** -- and **Exam 2**, -- both measuring the **same** **content**, -- to the **same group of students**.

Question 111

What is the Limitation of Parallel Forms Reliability?

Answer 111

Can be **quite time consuming**.

Question 112

What is Split Half Reliability aka?

Answer 112

1. **Cronbach’s alpha** 2. **Internal consistency** 2. **Reliability Across Items**

Question 113

What is Split Half Reliability?

Answer 113

The **reliability** -- of the **items** -- **within** a **measure**.

Question 114

What is a Huge Pro of Split Half Reliability?

Answer 114

Requires only **one assessment session**.

Question 115

Should you calculate split-half by correlating the first half of the test with the second half? Why or why not?

Answer 115

**No** — because of **practice effects** -- and -- **participant fatigue** affects **later items.**

Question 116

What is the assumption behind Split Half Reliability?

Answer 116

**Test scores** -- are **represented** -- by **total scores** -- across **all** the **items**

Question 117

How do you handle Split Half Reliability for tests measuring multiple constructs? -- Give example.

Answer 117

Scale which **assess** **multiple things** must **calculate** this **SEPERATELY** for **each construct**. * Example = **BFI-2**: ***Agreeableness*** and ***Neuroticism*** ***must*** be ***tested separately***.

Question 118

What is Interrater Reliability aka?

Answer 118

**Reliability Across Observers**

Question 119

What is Interrater Reliability?

Answer 119

The **reliability** -- of -- **two different observers** -- to **code** the **same thing**.

Question 120

What is the limitation of Interrater Reliability?

Answer 120

**1 observer** -- would **contain** -- too much -- **measurement error**!

Question 121

When does Interrater Reliability apply?

Answer 121

**Only** -- when **observations** -- are **being made** -- by **human raters**.

Question 122

What does Interrater Reliability require?

Answer 122

**Strict operational definitions**

Question 123

Give an example of-- **Strict operational definitions** -- for interrater reliability.

Answer 123

Clearly defining what **counts** as: * **Aggressive Behavior** * **Depression** * **Happiness**

Question 124

What is a common statistic used to measure Interrater Reliability?

Answer 124

**Cohen’s kappa**

Question 125

Read

Answer 125

Reliability is a special type of correlation **The higher the number, the better!**

Question 126

See picture

Answer 126

Question 127

**Validity** -- from here on--

Answer 127

-- from here on--

Question 128

What is Validity?

Answer 128

The **extent** -- which **your measure** -- is **measuring** -- * **What** it is **supposed to**, -- * **Has done so properly**, -- and * Is **generalizable**.

Question 129

What are the 3 kinds of Validity?

Answer 129

1. **Construct Validity** -- ***F C P - C C D*** a) ***Face*** b) ***Content*** c) ***Predictive*** d) ***Concurrent*** e) ***Convergent*** f) ***Discriminant*** 2. **Internal** **Validity** 3. **External** **Validity**

Question 130

What is Construct Validity?

Answer 130

The **extent** -- to **which** -- you are **measuring** -- what you are **supposed** -- to **measure**.

Question 131

Answer the following in other words: What is Construct Validity? Give Example.

Answer 131

**How good** -- are -- your **operational definitions**? Example: **How good** -- is the -- **operational definitions** -- of **Extraversion**?

Question 132

What are the 6 forms if Construct Validity?

Answer 132

1. **Face** **Validity** 1. **Content** **Validity** 1. **Predictive** **Validity** 1. **Concurrent** **Validity** 1. **Convergent** **Validity** 1. **Discriminant** **Validity**

Question 133

What is Face Validity?

Answer 133

The **extent** to **which** -- the **measure appears** -- to be **assessing** -- **what it’s supposed to**.

Question 134

How is Face Validity assessed?

Answer 134

**Assessment** based on -- **content** **of measure**.

Question 135

Face Validity = Worst type of validity

Answer 135

..

Question 136

Give 2 examples of items with Good Face Validity on a depression scale.

Answer 136

* **I feel sad.** * **I am depressed**.

Question 137

Give an example of item with Bad Face Validity on a depression scale.

Answer 137

* **I have low self-esteem.**

Question 138

Is face validity based on statistical evidence or subjective judgment?

Answer 138

**Subjective judgment** — it **involves** -- an **on-the-surface evaluation**.

Question 139

Why isn’t Face Validity a strong indicator of construct validity?

Answer 139

Because something -- that **looks right** -- may **not** **actually measure accurately** —— it **lacks objective evidence**.

Question 140

see picture.

Answer 140

Question 141

What is Content Validity?

Answer 141

**Does** -- your **measure** **include** -- **all** the **content** of the **construct** -- it is **measuring**.

Question 142

How is Content Validity assessed?

Answer 142

**Assessment** based on -- **content** **of measure**.

Question 143

Give an example of a measure with Good Content Validity for depression.

Answer 143

**Depression** measure -- with -- **all symptoms of depression**.

Question 144

Give an example of a measure with Poor Content Validity.

Answer 144

**Exam only on chapter 1** -- instead of **all 5 chapters**.

Question 145

What is Predictive Validity?

Answer 145

**Does** -- your **measure** -- **accurately** **predict** -- **FUTURE behaviors**?

Question 146

How is Predictive Validity assessed?

Answer 146

**Assessment** based -- on -- some **criterion variable**.

Question 147

Give 2 examples of Good Predictive Validity.

Answer 147

* **Measure** of **sensation seeking** -- predicting -- **willingness** to **go** **skydiving** * **GRE**-- predicting -- **success** in **grad school** (in theory…)

Question 148

Give 2 examples of Bad Predictive Validity.

Answer 148

* **Measure** of **Extraversion** -- **does not predict** -- **party attendance**. * **Measure** of **altruism** -- **does not predict** -- willingness to **help** a **stranger**.

Question 149

What is Concurrent Validity?

Answer 149

**Does** -- your **measure** -- **accurately predict behaviors** -- occurring **AT THE SAME** **time** **point**.

Question 150

How is Concurrent Validity assessed?

Answer 150

**Assessment** based on -- a **criterion variable**.

Question 151

Give 2 examples of Good Concurrent Validity.

Answer 151

Question 152

Give 2 examples of Bad Concurrent Validity.

Answer 152

Question 153

What is Convergent Validity?

Answer 153

**Does** -- your **measure** -- **correlate** -- with **other measures** of the **same** (or very similar) **things**?

Question 154

How is Convergent Validity assessed?

Answer 154

**Assessment** made -- using -- some **criterion variable**.

Question 155

Give an example of Good Convergent Validity.

Answer 155

✅ Example: **Measuring Depression** Let’s say a researcher wants to assess depression and uses two different tools: Beck Depression Inventory (**BDI**) Patient Health Questionnaire-9 (**PHQ-9**) **Both are well-established self-report scales for measuring depression**.

Question 156

Give an example of Bad Convergent Validity.

Answer 156

Question 157

What is Discriminant Validity?

Answer 157

**Is** -- your **measure** -- **unrelated** -- to **things** -- it **should** **not** -- be **related** **to**?

Question 158

How is Discriminant Validity assessed?

Answer 158

**Assessment** -- made using -- some **criterion variable**.

Question 159

Why is discriminant validity sometimes left out of research?

Answer 159

Because -- it is -- **often** **difficult** -- to **assess** -- **effectively**.

Question 160

What is one reason to assess discriminant validity in new scales?

Answer 160

To show the **benefits** of **additions** -- or -- **distinctions** -- made in **new measurement scales**.

Question 161

Give an example of discriminant validity from emotional research.

Answer 161

**Valence** (***positive*** or ***negative emotion***) -- is often **shown** -- to be **distinct** -- **from** **arousal**.

Question 162

How does the Shy Q demonstrate discriminant validity?

Answer 162

The **Shy Q** -- is **not** **correlated** -- with **interpersonal** **forcefulness**, -- showing it's **measuring** -- a **distinct** **interpersonal trait**.

Question 163

Why is the Shy Q vs. interpersonal forcefulness a good example of discriminant validity?

Answer 163

Because **both** are **interpersonal variables, --** yet they **remain uncorrelated**, -- supporting the **distinction** between **constructs**.

Question 164

What is **Internal Validity**?

Answer 164

The **degree** -- to which **causal** **conclusions** -- can be **drawn** -- from the **results** -- of the **study**.

Question 165

What key question does internal validity address regarding causality?

Answer 165

1. **Did** you **establish** the **conditions** for **causality**? 1. **How** well **did you control** for potential **third** **variables**?

Question 166

What is External Validity?

Answer 166

The **degree** -- to **which** -- **results** are **generalizable** -- to **other** **populations** and the **real world**.

Question 167

What key question does External Validity address?

Answer 167

1. **How** well -- **do** the **lab results** -- translate to **real life**? 1. Often, we **sacrifice external validity** to **gain internal validity**, or **vice versa**. -- (By increasing internal validity, you decrease external validity, and vice versa).

Question 168

see picture.

Answer 168

Question 169

See picture

Answer 169

Question 170

see picture

Answer 170

Question 171

**ADDITIONAL CONSIDERATIONS** -- from here on--

Answer 171

-- from here on--

Question 172

What are Response Sets?

Answer 172

A **tendency** for **individuals** -- to **respond** -- to **questions** -- **on** a **basis** -- that is **unrelated** to **question content**.

Question 173

Give 3 example of Response Sets.

Answer 173

1. **Acquiescence** 2. **Extreme Responding** 3. **Social Desirability**

Question 174

What is Acquiescence in Response Sets?

Answer 174

The **tendency** -- for **individuals** -- to **agree** -- with **ALL items**.

Question 175

What is a Solution for Acquiescence?

Answer 175

**Reverse Scored Items**

Question 176

What is Extreme Responsing in Response Sets?

Answer 176

**Tendency** -- for **individuals** -- **to give** -- **end-point responses**. (selecting ***most strongly agree*** or ***strongly disagree items***).

Question 177

What is a solution for Extreme Responsing?

Answer 177

**Attention Checks**

Question 178

What is Social Desirability in Response Sets?

Answer 178

The **tendency** -- **to** **respond** -- to **items** -- **in** a **way** -- that **makes** the **participant** seem **likeable**.

Question 179

Is Social Desirability the same as lying?

Answer 179

**No**—it is not necessarily lying or faking, -- but a **bias** **toward** **appearing favorable**.

Question 180

What is a solution for handling Social Desirability Bias?

Answer 180

**Measure social desirability** -- and -- **control** for it **statistically**.

Question 181

See picture

Answer 181

Question 182

**Sources of Data - LOTS** -- from here on--

Answer 182

-- from here on--

Question 183

What are the 4 main Sources of Data in Personality Psychology?

Answer 183

1. **Life Data** 2. **Observer Data** 3. **Test Data** 4. **Self Data**

Question 184

What is Life Data?

Answer 184

**Refers** to -- any **obtainable** **record** -- of the **individual’s life** -- which is **objective** -- and -- **publicly available**.

Question 185

What are examples of Life Data (L-data)?

Answer 185

1. **Criminal Record** 2. **Birth Date** 3. **Medical Record** (***consent required***)

Question 186

How is L-data often used in research?

Answer 186

It is often -- **predicted** by **other sources**, -- such as **personality traits**. * Example: does ***self-reported Agreeableness*** lead to ***marrying earlier***?

Question 187

Can L-data be used in real time?

Answer 187

**Yes**. * In-the-moment **observations** can be **made**, -- like ***insurance companies*** -- using ***driving records*** -- to ***adjust*** ***rates***.

Question 188

What is an Advantage of Life-data?

Answer 188

**Objectivity**

Question 189

Give an example of Advantage (Objectivity) of Life Data.

Answer 189

**Fraternity drinking behavior**. * President of **university** = **paid** someone to go to **garbage** -- and -- **see** the **#** of **alcohol beverages**. -- **People reported much less than** they **drank**).

Question 190

What is a disadvantage of Life-data?

Answer 190

1. **Data** can be **hard**/**time consuming** -- to **collect**. 1. Only **applicable** -- to -- **certain research questions**.

Question 191

What is Observer Data?

Answer 191

Refers to -- any **data** **collected** -- **using** -- an **independent observer**

Question 192

Why is fraternity garbage an example of L-data and not O-data?

Answer 192

* Because the **observer** is **not** **coding behavior**, just collecting **objective data** (e.g., ***counting bottles***). * This O-Data requires **observer** -- **TO** **CODE** **behavior** -- or **reporting** -- the **data themselves**.

Question 193

Give 2 examples of Observer-data?

Answer 193

* **Friend** -- **reporting** -- on your **extraversion**. * **Research assistant** -- **coding** -- for **authenticity** in a **social interaction**.

Question 194

What is one Advantage of Observer-data?

Answer 194

1. **Solves** **issues** -- of **desirability bias** in S data (somewhat…) 1. Sometimes -- **other people know** -you **better** -- than -- you **know yourself** 1. **Multiple observers** -- **can** be -- **used**. a) ***Information*** -- is ***more*** ***reliable***.

Question 195

What is one Disadvantage of Observer-data? - Give example.

Answer 195

**Observer subjectivity** * Example: ***Agreeableness*** = ***how great you are*** at ***maintaining social relationships***.

Question 196

What is Test-data?

Answer 196

**Refers** -- to any **objectivity data** -- **obtained** -- **via standardized testing**.

Question 197

Give 2 examples of test-data.

Answer 197

* **GRE** * **LSAT**

Question 198

What are Advantages of Test-data?

Answer 198

1. **Data** is **objective**. 1. Can **elicit** -- **difficult** to **observe behaviors** a) Example: ***Leadership roles***. 1. **Allow** -- **control** -- over **confounding variables**.

Question 199

What are Disadvantages of Test-data?

Answer 199

1. **Participants** may **guess** -- **what** is **being measured**. 1. **Subject** to **experimenter effects** (***sometimes***). ## Footnote experimenter effects = unintended influences a researcher has on the results of their study, either consciously or unconsciously.

Question 200

What is Self-data?

Answer 200

Refers -- to **any** **data** -- **collected** -- **via self report**.

Question 201

What belongs in Self-data category?

Answer 201

**Almost all surveys**

Question 202

What types of questions are used in Self-data?

Answer 202

1. **Unstructured** 2. **Structured**

Question 203

What are Unstructured questions? -- Give example.

Answer 203

**Open-ended questions**. * Examplease: ***Tell me about yourself***.

Question 204

What are Structured questions? -- Give example.

Answer 204

The **questions** -- that **impose** -- **some sort** -- of **response scale**. * Example: ***How extraverted are you*** (***1-5***)?

Question 205

What are Advantages of Self-data?

Answer 205

1. **Easiest** -- **type of data** -- to **collect**. 1. **Allows** -- for -- **large sample size**.

Question 206

What are Disadvantages of Self-data?

Answer 206

1. **Sampling bias** 1. **Social desirability bias** 1. **Sensitive topics** -- (e.g., ***sexual behavior***) -- can **generate** -- lots of -- **missing data**.

Question 207

Answer 207