the statistical model Flashcards

1
Q

All models we use are wrong

A

“All models are wrong, but some are useful” (Box, 1976)

  • They are representations of a thing that do not necessarily capture all the complexity of reality
  • But they can still be of use to us
  • E.g. the underground map or the model of a car’s suspension
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

The statistical model

A
  • We don’t make physical models of things
  • Small bridges, maps
  • Our models are made from data
  • We collect data about a phenomenon
  • Look for patterns or trends
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Simple model

A

Describing central tendency and variability in a measure

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Complex model

A

Network of how different variables connect to each other

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

step 1

A

We find a phenomenon we are interested in

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

step 2

A

We create a theoretical model & create study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

step 3

A

We collect data about that phenomenon

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

step 4

A

Visualise & interpret the variability in that data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

step 5

A

Describe the variation using appropriate descriptive statistics

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

step 6

A
  • This is where we build and interpret the statistical model (in RStudio)
  • Based on our theoretical model and using our data
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

• Why do we need a statistical model?

A
  1. Make conclusions about lots of people by just studying a few
  2. Hypothesis testing
  3. Estimation
  4. Predict the future
  5. Compare different theoretical models
How well did you know this?
1
Not at all
2
3
4
5
Perfectly