Nayland College

Nayland College - Mathematics

Home . Year 9 . Year 10 . Level 1 . Level 2 . L3 Statistics . L3 Maths . L3 Calculus . About . Links

NZAMT NZQA NZ Grapher NZ Maths Census at School Study It

3.10 Inference HOME | Achievement Objectives | Overview | Statistical Cycle | Introduction & Research Question | Discussing Samples | Sample Variability | Sampling Methods | Informal Confidence Interval | Bootstrapping Confidence Interval | Comparing two populations | Using iNZight | Writing a report | Assessment Practice

iNZight & Inference

Remember: You Have to run iNZight TWO times!

The iNZight tips page has information on ordering, removing, grouping, re-categorising, & exporting data, and creating variables, adding variables to create box plots and show summary statistics. (Adding another variable can also provide more information)

***Manipulating data in Excel first can be a good idea, especially to combine groups together notes Class notes

 

1) To Analyse the Sample

2) To make a Bootstrap Confidence Interval

Ensure you have the latest version of iNZight downloaded

Link to iNZight site free download and videos on how to get started 

Run iNZight

1) To Analyse the Sample

Run iNZightRun iNZight normally, import data

Decide on your variable, and what your two groups are (see iNZight tips page for ordering, removing, grouping, re-categorising, & exporting data, and creating variables)

Add variables to create box plots and show summary statistics. (Adding another variable can also provide more information (iNZight tips page)

Input Data

Input Data Not sure how to Import Data? Find out here...

REMEMBER: If you upload your own data it MUST be in a .csv file, with variable notation, no commas, no variable definitions etc

Make a dot plot & Box plot

Input Data Drag ONE VARIABLE

and

ONE Group to the 'variables'

Note: You may want to manipulate the data in excel first, eg If therre are multiple groups thencombining some together so there is only two groups would be a good idea.

 

Input Data

Get SAMPLE SUMMARY statistics

Input Data

 

Input Data

 

 

 

Now DISCUSS the SAMPLE STATISTICS & DOT plot & BOX plot
Remember this by the starship USSCSI or 'DISCUSS' (see the sample page if you are not sure)

 

2) Bootstrapping using iNZight

Difference between two Population Groups

iNZight Bootstrap confidence intervals

How to compare two groups and answer "Is the median on Blah different from the median of Blah in the population of Blah?

Import data which has two categories

The two columns of data - one is measured data and the other is the group

Add variables

(variable 1 is primary variable, Variable 2 is group)

Analyse

Discuss the dot plots for the sample groups.

The Arrow represents the difference between the medians of the two sample groups

Remembering that the samples vary each time.

Does the population medians also differ?

Analyse, record my choices

The differences between the pairs of data values is the 'sample'

Bootstrap resamples are taken from this 'sample' of differences.

 

Resample

You can watch one animation to get the idea..

or resample 1000 times to see the variation in the resampled medians.

The top dot plot is the original sample differences

The bottom dot plot is of a single resample of size 'n' with replacement.

The differences in the resample medians is recorded

 

Bootstrap Distribution

You can watch one animation to get the idea..

then Re-sample 1000 times to see the bootstrap distribution of the differences between repeated re-samples from each group

Show the CI to see the Bootstrap confidence interval.

Interpret what you see

Multiple re-samples build up a distribution of differences between the means

(see the compare populations page if you are not sure)

How do we interpret these results?

What does this tell us?

You MUST discuss the relative position of 0 and what this means with respect to the two population medians (or means)

 

 

 

 

back to top