A “Short” Introduction to Experimental and Quasi-Experimental Designs

class: center, middle, inverse, title-slide

# A “Short” Introduction to Experimental and Quasi-Experimental Designs
## EDP 612 Week 5
### Dr. Abhik Roy

---

section p {
    margin: auto;
}

section {
    height: 600px;
    width: 60%;
    margin: auto;
    border-radius: 21px;
    background-color: #212121;
}

section p {
    text-align: center;
    font-size: 30px;
    background-color: #212121;
    border-radius: 21px;
    font-family: Roboto Condensed;
    font-style: bold;
    padding: 12px;
    color: #bff4ee;
}

#center {

text-align: center;

}

.center p {
  margin: 0;
  position: absolute;
  top: 50%;
  left: 50%;
  -ms-transform: translate(-50%, -50%);
  transform: translate(-50%, -50%);
}

.center2 {
  margin: 0;
  position: absolute;
  top: 50%;
  left: 50%;
  -ms-transform: translate(-50%, -50%);
  transform: translate(-50%, -50%);
}

.tab {
    display: inline-block;
    margin-left: 40px;
}
</style>

<div>
<style type="text/css">.xaringan-extra-logo {
width: 110px;
height: 128px;
z-index: 0;
background-image: url(/Users/skynet/Documents/WVU/Teaching/GitHub.nosync/edp612/static/img/course_hex.png);
background-size: contain;
background-repeat: no-repeat;
position: absolute;
top:1em;right:1em;
}
</style>
<script>(function () {
  let tries = 0
  function addLogo () {
    if (typeof slideshow === 'undefined') {
      tries += 1
      if (tries < 10) {
        setTimeout(addLogo, 100)
      }
    } else {
      document.querySelectorAll('.remark-slide-content:not(.title-slide):not(.inverse):not(.hide_logo)')
        .forEach(function (slide) {
          const logo = document.createElement('a')
          logo.classList = 'xaringan-extra-logo'
          logo.href = 'https://edp612.asocialdatascientist.com'
          slide.appendChild(logo)
        })
    }
  }
  document.addEventListener('DOMContentLoaded', addLogo)
})()</script>
</div>

---

class: highlight-last-item
layout: true

---

# A Typical List of Variable Types

> There are just so many!

<br>
> You may need to memorize these in the future but for now, this is a not very comprehensive list

<br>
> Mostly jargon free because communication is more important than sounding smart

<br>
> Popular falesly attributed quotes but still apply:

>-  *You do not really understand something unless you can explain it to your grandmother*
  
  >- *If you can't explain it simply, you don't understand it well enough*
  
  >- *Explain it to me like I'm an idiot*

---

## The Basic Comparisons

.pull-left[ 
<span style="color:#729ec6;">Independent</span>
<br>
the cause
]

<br>
.pull-right[ 
<span style="color:#faa1a1;">Dependent</span>
<br>
the effect
]

<br>
.pull-left[ 
<span style="color:#729ec6;">Explanatory</span>
<br>
what is changed
]

<br>
.pull-right[ 
<span style="color:#faa1a1;">Response</span>
<br>
what happens from that change
]

<br>
.pull-left[ 
<span style="color:#729ec6;">Mediator</span>
<br>
something that can somewhat or fully explain cause and effect
]

<br>
.pull-right[ 
<span style="color:#faa1a1;">Moderator</span>
<br>
something that impacts the direction and/or strength between cause and effect 
]

---

## The Basic Ones Without Good Comparisons

<center>
<span style="color:#b6d7a8;">Extraneous</span>
<br>
<br>
something known but not being studied that could influence cause and effect
</center>

<br>
<center>
<span style="color:#b6d7a8;">Confounding</span>
<br>
<br>
something not known that could influence cause and effect
</center>

<br>
<center>
<span style="color:#b6d7a8;">Control</span>
<br>
<br>
a baseline that doesn't change
</center>

---

# Part I: Nonexperimental Research

- Lacks the manipulation of an independent variable

- While we cannot show causation, this type of research is still very important
		
---

##  When Should We Use Non-Experimental Research?

-  You only have a single variable

-  You have a non-causal relationship between variables

- There is a causal relationship, but you can’t (practically or ethically) randomly assign participants

- There's a broad or exploratory question that needs answered

- Need to make sense of data

- *Examples*

> *Does damage to a person’s hippocampus impair the formation of long-term memory traces?*
  
  > *Is there some relationship between verbal intelligence and mathematical intelligence? *
  
  > *How accurate are people’s first impressions?*
  
  > *What is it like to be a working mother diagnosed with depression?*

---

## Types: Correlational Research

---

###  What is Correlational Research?

- Non-experimental research design used to assess the statistical relationship between two variables

- Little to no effort is used to account for confounding variables

---

###  Why Do Correlation Research?

- If there’s no correlation, then there can’t be eventual causation

- Ideas for future experiments

- Some studies are too impractical or unethical to be done using an experiment

---

###  Goals for Correlation

1. *Descriptive*.

- Describes relationships
  
--
  
  - *Example*. “Does bragging about wealth correlate with scores on the [Narcissistic Personality Inventory](https://openpsychometrics.org/tests/NPI/)?”

2. *Predictive*.

- Predict behavior
  
--

- *Example*. “If a rich person scores 34 (greater level of narcissism) on the Narcissistic Personality Inventory (predictor variable), can we can predict that individual is more likely to brag about wealth (outcome variable)?”

---

###  Maybe / Could Be vs Probably / Certainly

.pull-left[ 
<center>
<span style="color:#faa1a1;">Correlational</span>
<hr style="width:50%;">
</center>
]

.pull-right[ 
<center>
<span style="color:#729ec6;">Experimental</span>
<hr style="width:50%;">
</center>
]

<br>
.pull-left[ 
<center>
<span style="color:#fdd1d1;">Cannot determine cause</span>
</center>
]

.pull-right[ 
<center>
<span style="color:#a8c3dc;">Causal</span>
</center>
]

<br>
.pull-left[
<center>
<span style="color:#fdd1d1;">Cannot control variables</span>
</center>
]

.pull-right[
<center>
<span style="color:#a8c3dc;">Experimental control</span>
</center>
]

<br>
.pull-left[
<center>
<span style="color:#fdd1d1;">Observe and describe possible relationships</span>
</center>
]

.pull-right[
<center>
<span style="color:#a8c3dc;">Manipulate and describe changes</span>
</center>
]

<br>
.pull-left[
<center>
<span style="color:#fdd1d1;">Find individual differences</span>
</center>
]

.pull-right[ 
<center>
<span style="color:#a8c3dc;">Find group differences</span>
</center>
]

---

### Sometimes Making Sense Out a Mess

---

# Calendar of Tasks

<center>
<table class="table" style="width: auto !important; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:center;">  </th>
   <th style="text-align:left;">  </th>
   <th style="text-align:left;">  </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L0 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L0: Lit Review Process </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 3 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L1 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L1: Topic Selection &amp; Annotated Bib </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 5 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L2 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L2: Annotated Bibliography </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 7 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L3 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L3: Lit Review First Draft </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 9 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L4 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L4:  Lit Review Second Draft </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 11 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L5 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L5: Lit Review Third Draft </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 13 </td>
  </tr>
  <tr>
   <td style="text-align:center;width: 5em; color: #ffffff !important;vertical-align: middle !important;"> L6 </td>
   <td style="text-align:left;width: 20em; color: #ffffff !important;vertical-align: middle !important;"> L6: Lit Review Final Version! </td>
   <td style="text-align:left;width: 10em; color: #ffffff !important;vertical-align: middle !important;"> Week 15 <sup>1</sup> </td>
  </tr>
</tbody>
</table>
</center>

<a style="font-size:14px;"><sup>1</sup>This is the week of final exams.</a>

---

# Part II: Experimental Design

---

#  Two Features of an Experiment

+ Manipulation of the Independent Variable

+ Can “cause” this independent variable to occur
	
--

+ This “cause” elicits the response (depenedent variable)
	
--

+ Random Sampling
	
--

+ Control of Extraneous Variables
	
--

+ Can hold variable(s) constant
	
--

+ Limit variable to a specific category
	
--

+ (Random Assignment)
	
--

*Quasi Experiment* - Can typically manipulate the IV, but cannot randomly assign participants to conditions

---

#  Types of Treatment & Control Groups

+ Randomized clinical trial

+ One group gets the “treatment” and another group doesn’t –the control
	
--

+ No-treatment control condition	
--

+ Zero Treatment (they get nothing)	
--

+ Problem: placebo effect	
--

+ Placebo Control group	
--

+ Given Placebo	
--

+ Checks for placebo effect	
--

+ Wait-list control condition	
--

+ Delayed treatment	
--

+ Checks for “expectation of improving”

---

## Random vs. Random
 
--

.pull-left[ 
<center>
<span style="color:#faa1a1;">Random Sampling</span>
<hr style="width:50%;">
</center>
]

.pull-right[ 
<center>
<span style="color:#729ec6;">Random Assignment</span>
<hr style="width:50%;">
</center>
]