Sampling

class: center, middle, inverse, title-slide

.title[
# Sampling
]
.subtitle[
## EDP 619 Week 3
]
.author[
### Dr. Abhik Roy
]

---

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

.remark-slide-container {
background: #212121;
}

.hljs-github .hljs {
    background: transparent;
    color: #b2dfdb;
}

.hljs-github .hljs-keyword {
    color: #64b5f6;
}

.hljs-github .hljs-literal {
    color: #64b5f6;
}

.hljs-github .hljs-number {
    color: #64b5f6;
}

.hljs-github .hljs-string {
    color: #b7b3ef;
}

.hljs-github .hljs {
    background: transparent;
    color: #b2dfdb;
}

.hljs-github .hljs-keyword {
    color: #64b5f6;
}

.hljs-github .hljs-literal {
    color: #64b5f6;
}

.hljs-github .hljs-number {
    color: #64b5f6;
}

.hljs-github .hljs-string {
    color: #b7b3ef;
}

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;
    margin: auto;
}

#center {
text-align: center;
}

#right {
  text-align: right;
}

.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/edp619/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://edp619.asocialdatascientist.com'
 slide.appendChild(logo)
 })
 }
 }
 document.addEventListener('DOMContentLoaded', addLogo)
})()</script>
</div>

---

class: highlight-last-item
layout: true

---

# Two Approaches to Sampling

.pull-left[

Nonprobability 
each person in your target population does not have an equal chance of being selected

]

.pull-right[

Probability 
each unit in your target population must have an equal chance of being selected

]

---

# Nonprobability Sampling

>- Probability is usually unknown

>- Does not rely on numerical data

>- Inability to generalize to any populous

>- Used when you want to say something about a discrete phenomena, a few select cases (people, places, objects, etc)

---

## Characteristics

> Easier than probability based methods

> Nonrandom selection

> Sampling bias is present

> Samples are not considered representative of the populations from which they were drawn

---

## Primary Types

.pull-left[

Convenience

]

.pull-right[

Purposive

]

.pull-left[

Quota

]

.pull-right[

Snowball

]

---

## Convenience Sampling1

.pull-left[
* Samples are selected based on

> their availability to the researcher

* Good for

> administering a pilot study
  
  > generating a hypothesis
  
  > gaining an initial sense of attitudes or opinions
]

.pull-right[
<center>
 

Example 
Crowdsourcing survey participants from a platform2

 
</center>
]

.footnote[1 aka **haphazard** or **accidental** sampling 
2 like 
<a href="https://www.mturk.com/" target="_blank">Amazon Mechanical Turk (MTurk)</a>]

---

## Purposive Sampling

.pull-left[
* Samples are selected based on

> elective criteria that define a unique group
 
 > targeting knowledgeable individuals1

* Good for

> focusing on the depth of relatively small samples
  
  > identifying cases, individuals, or communities best suited for a study
  
]

.pull-right[
<center>
 

Example 
Choosing skilled candidates for a job vacancy

 
</center>
]

.footnote[1 aka **key informants**]

---

## Quota Sampling

.pull-left[

* Samples are selected based on

> defined subgroups that exhibit certain characteristics of interest

* Good for

> gaining insight about a characteristic of a particular subgroup
  
  > investigating relationships between different subgroups
  
]

.pull-right[
<center>
 

Example 
Assessing the the differences in the career goals among university freshman, sophomores, juniors, and seniors

 
</center>
]

---

## Snowball Sampling1

.pull-left[

* Samples are selected based on

> individuals recruited by other individuals

* Good for

> researching people with specific traits who might otherwise be difficult to identify and/or gain access to
  
  > keeping costs low
  
]

.pull-right[
<center>
 

Example 
Studying the current living status of ex-convicts

 
</center>
]

.footnote[1 aka **chain** or **network** sampling]

---

## Why should I even care?

--
 
 
Because:

> Any choice will limit the type of utilizable quantitative study

> Not everything can be explained quantitatively

> Some studies even mandate a mixed methods design

---

# Probability Sampling

> Based solely on the idea that a population can be represented by a subset of it given some error: **Random selection**

<center>
Example: 45% ± 3% agree with...
</center>

> Ability to generalize to a certain populous

> Inability to describe individual phenomena at any great depth

---

## Characteristics

> Greater difficulty than non-probability based methods

> Random selection

> Sampling bias is minimal, and samples are considered representative of the populations from which they were drawn

> Samples are representative of the populations from which they were drawn

---

## Primary Types

.pull-left[

Census

]

.pull-right[

Simple Random Sample (SRS)

]

<center>
 

Systematic

 
</center>

.pull-left[

Stratified

]

.pull-right[

Cluster

]

---

## Census

.pull-left[

* Samples are selected based on

> an official count or survey of a population, typically recording various details of individuals

* Good for

> ease of administration
  
  > generalizing to an overall populous
  
  > simple data analysis
  
  > small samples
  
]

.pull-right[
<center>
 
 
Example 
The United States Census

 
</center>
]

---

### General Idea

---

## Characteristics

.pull-left[
Benefits
  > a lack of an error associated with a result
  
  > self-weighting

]

.pull-right[
Drawbacks
  > extremely expensive
  
  > time consuming
  
  > typically infeasible
]

---

## Simple Random Sample (SRS)

.pull-left[

* Samples are selected based on

> an equal probability of being picked

* Good for

> ease of administration
  
  > generalizing to an overall populous
  
  > simple data analysis
  
  > situations where not a lot is known about a population
  
  > large samples
  
]

.pull-right[
<center>
 
 
Example 
Drawing names from a hat

 
</center>
]

---

## Characteristics

.pull-left[
Benefits

> data collection can be efficiently performed on randomly distributed items
  
  > simple error calculation
  
  > self-weighting

]

.pull-right[
Drawbacks
  > expensive
  
  > likely impractical
  
  > possible underrepresentation of subgroups
  
  > tedious
  
  > time consuming
  
  > vulnerable to sampling errors
]

---

### General Idea

---

## Systematic Sample

.pull-left[

* Samples are selected based on

> arranging of a population according to some ordering pattern and then the selection of elements at regular intervals from that that ordered list

* Good for

> ease of administration
 
 > automation of selection process1
 
 > providing more information about a population than an SRS
 
]

.pull-right[
<center>
 
 
Example 
Picking every third house on a block to poll

 
</center>
]

.footnote[1 after selecting the first unit]

---

## Characteristics

.pull-left[
Benefits

> most likely will provide a more robust information set per unit cost than an SRS
  
  > less subjective to selection error than SRS
  
  > simple selection process
  
]

.pull-right[
Drawbacks
  > dependence on a previous and next unit
  
  > vulnerable to periodicities

]

---

### General Idea

---

## Stratified Random Sampling

.pull-left[

* Samples are selected based on

> a population being divided and subdivided into distinct groups1 followed by a simple random or systematic sample in each

* Good for

> ease of administration
 
 > automation of selection process1
 
 > providing more information about a population than an SRS
 
]

.pull-right[
<center>
 
 
Example 
Administering a survey to random units of all apartment complexes in a town

 
</center>
]

.footnote[1 aka **strata**]

---

## Characteristics

.pull-left[
Benefits

> less variability than an SRS
  
  > reduced sampling error
  
  > reduced reported error and increases precision compared to an SRS
  
]

.pull-right[
Drawbacks
  > may be expensive
  
  > strata must be implicitly defined

]

---

### General Idea

---

## Cluster Random Sampling

.pull-left[

* Samples are selected based on

> a population being divided and subdivided into distinct groups1 followed by a random sample of those units with census in each

* Good for

> when lacking a sampling frame
  
  > cost efficiency is needed
  
]

.pull-right[
<center>
 
 
Example 
Picking every third house on a block to poll

 
</center>
]

.footnote[1 aka **clusters**]

---

## Characteristics

.pull-left[
Benefits

> clusters can be stratified if necessary which results in increased precision
  
  > less subjective to selection error than SRS
  
  > simple selection process
  
]

.pull-right[
Drawbacks
  > may not represent diversity within a populous
  
  > prone to high sampling errors
  
  > requires a larger sample size than SRS

]

---

### General Idea

---

## That's it!

If you have any questions, please reach out

<center>
<a rel="license" href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img alt="Creative Commons License" style="border-width:0" src="https://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png" /></a> This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>
</center>