Lab in R programming
Falcon need do this lab using the info in ppt and do it in R programming
2 years ago
10
files (1)RESEV552Week12FinishinLM2.pptx
RESEV552Week12FinishinLM2.pptx
Finishing the Hat: Linear Regression from start to finish and maybe some network analysis
RESEV 552
Agenda
Working through our linear models
Checking our work
Thinking about implications for policy and practice
Network Analysis Basics
Today
Check your data set
Identify the distribution of your variables.
Write our your guiding research question.
Write out your hypotheses.
Conduct your hypothesis test.
Interpret your results
Develop implications for policy and practice
Network basics
Check your data
What variables are available?
What is the 5-number summary for continuous variables. Make a table and put this table in your manuscript.
If you have binary categorical variables, what is the distribution of proportions?
Check for NAs
Check for potential coding errors
Identify the distribution of your variables
Give me a meaningful graph and interpret it for the reader given what you know about distributions.
Write your guiding research question
Remember, your RQ can contain multiple variables, but your hypothesis test cannot.
If you are going to do multiple hypotheses tests, ask a question that’s broad enough to contain each hypothesis.
Good Example: What institutional and individual influences are significantly related to changes in proportion of students who graduate in five years?
Bad Example: Do higher levels of wealth result in students graduating in five years?
Write your hypothesis statement
Ho: There is not significant relationship between school funding and graduation rates.
Ha: School funding is significantly related to graduation rates.
Hypothesis for the slope of the partial relationship between x and y, controlling for w and d:
H0 yx●wd : β1 = 0
HA yx●wd: β 1 ≠ 0 OR β 1 > 0 OR β 1 < 0 [Note: Select only one of these]
Hypothesis for the slope of the partial relationship between w and y, controlling for x and d:
H0 yw●xd : β 2 = 0
HA yw●xd: β 2 ≠ 0 OR β 2 > 0 OR β 2 < 0 [Note: Select only one of these]
Hypothesis for the slope of the partial relationship between d and y, controlling for x and w:
H0 yd●xw: β 3 = 0
HA yd●xw: β 3 ≠ 0 OR β 3 > 0 OR β 3 < 0 [Note: Select only one of these]
Conduct your hypothesis test
In this sample, the estimated slope of the partial linear relationship between {independent variable x} and {dependent variable y}, after controlling for {independent variable w} and {independent variable d}, is {b1}.
2. If the true slope of the partial linear relationship between {independent variable x} and {dependent variable y} among {focal population}, after controlling for {independent variable w} and {independent variable d}, is zero, the probability of obtaining a random sample of size {n} with a partial linear relationship {that differs from zero by as much or more than OR as large or larger in absolute magnitude and in the same direction as} the observed sample partial linear relationship of {b1 } is {calculated p}.
This calculated probability is {greater than OR less than} the α of {?}.
Thus, I {can OR cannot} reject the plausibility of zero for the true slope of the partial linear relationship between {independent variable x} and {dependent variable y} among {focal population}, after controlling for {independent variable w} and {independent variable d}.
Interpret your results, implications
What does your hypothesis test suggest based on the design of your study?
What changes for policy or practice, if any, should be considered based on your research?
Network Basics
Individual (Ego)
Connections (Tie)
Other people (Alters)
Organizational or Social Groups (Network Boundary)
Before Exam 1
Before Exam 3
Men Women
11
02
Research Design
Gender connections
11
Core Discussion Network
“From time to time, most people discuss important matters with other people. Looking back over the last six months—who are the people with whom you discussed matters important to you?”
Small, M. L. (2013). Weak ties and the core discussion network: Why people regularly discuss important matters with unimportant alters. Social networks, 35(3), 470-483.
Next:
Place yourself in the center and your connections around you
Draw connections to yourself and to other individuals in your network who know each other
Take a second
Who is present in your advice seeking network?
Who is well connected to other folks? Disconnected?
What kind of benefits do you get from these relationships (informational, emotional support, financial or other resources)?
Who might you want to add to your network that’s missing?
How to quantify connections-Edge List
| From | To |
| 1 | 2 |
| 1 | 3 |
| 1 | 4 |
| 1 | 5 |
| 2 | 3 |
| 2 | 5 |
| 3 | 4 |
Quantifying community
Network size
Density-(Total number of ties * (Total number of ties-1))/2
Centrality- take every pair of the network and count how many times a node can interrupt the shortest paths (geodesic distance) between the two nodes of the pair
Diameter- longest distance across the network
Efficiency- Capacity for the network to remain undisrupted when individuals (nodes) are removed
