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

Using iNZight Tips & Tricks

word docs Word 'handout' version of using iNZight

Using Run iNZight:

iNZight free download and videos on how to get started (link to stat.auckland)

Manipulate Data

Filter data in INZight

Order data in iNZight

Manipulate variables

Convert to categorical Transform Variables Collapse levels Rename Levels ombine categorical variables Create New variables Reshape dataset Class intervals Rename variables Standardise variables


- convert, (numerical values to categorical values)
- transform, (eg log of...)
- collapse levels (combine groups together),
- re-name levels (groups),
- re-shape, (reorganise a spreadsheet)
- combine variables, (put categories together)
- create variables,
- class intervals, (create groups from a numerical data values),
- re-name variables,
- standardise (to a standard normal distribution)

Time series

Data Displays

- comparing groups,
- adding another group or variable to a dot plot

Bootstrapping Randomisation Using iNZight Visual Inference Tools:

(click on VIT to jump to tips)


back to top


Ordering data

Data IN/OUT as per normal, then right click on the column heading to:
'Apply a function to a column' sort by column decreasing or increasing, or 'rename the column'


Filtering data - 'select cases'

There are several options are available here:

Remember if you stuff it up you can always 'filter data' - 'restore data set'

'Levels of a categorical variable'
Good for situations where there are several groups in a category and you only want 2 groups to investigate.

'Numerical condition'
Good for situations where there you want to remove some data such as outliers or for some reason (eg under age drinkers) from a numerical column.

Remember you select data you want to keep. 


'Row number'
Good for removing specific data or outliers.
eg. Order the data by a column, view the data and note the row number, then filter the data by that row number to remove the outlier.

Good for testing the robustness of time series model

back to top

Data display in iNZight


Drag Variable 1 and Variable 2 (usually one numerical and one categorical)

Add in 'subset by' then move the slider bar to quickly flick between the different combinations of the category and subgroup (subset)

Great to do when initially investigating data sets. "I wonder if..."

Add to plot in iNZight

remember you can always 'remove additions' to return to basic plot

Code more variables adds colouring to the dot plot to show another variable.

Also great to do when initially investigating data sets. "I wonder if..."

Change plot appearance to reduce size of dot plot points reduces the 'stacking tilt' which we sometimes see in the dot plot.


back to top

Manipulate Variables

Some great options here. Explore.

Remember if it all goes wrong, just re-import your data again



'Convert to categorical'

By default iNZight takes numerical data to be numerical, not categorical.

If you have data that has categories identified numerically you can convert them to 'text'.

This will be added to the data beside the numerical column (good idea to re-name it)


'Transform variables'

Useful when we are modelling, Bivariate or looking at multiplicative models in time series.


‘Collapse levels'
When you want to combine groups within a category.

eg Qualifications:
we want to group ‘degree’ & ‘vocational’ together, and compare with ‘none’ & ‘school’ combined together.

Select variable, select groups to combine, ‘collapse, then repeat for other group, ‘collapse’ then ‘all done’

Another column will be added into the data set with the new groups. Useful for inference when we compare only two groups.

‘Re-Name levels'
When you want to re-name groups (levels) within a category.

Select your variable, then rename.

Good if you have converted numerical to categorical and want to name groups more appropriately.

'Reshape the data'

If you want to reorganise your data set without having to use Excel.


'Combine Categorical Variables'

This is when  you want to combine several categories together.


'Create New Variables'

Good for excellence in time series eg spending per person.

Be very careful you get the variable names correctly typed.

The best way is to ‘click and drag’ the variables into the ‘new variable’ box (from the 'view variables' list - not the 'data view set'

'Form Class Intervals'

Handy when you want to create groups from a numerical column. Lots of choice over number of intervals and the method.

eg. splitting into age groups.

'Re-name Variables'

Pretty obvious what this does.

'Standardise Variables'

This converts a variable into a standard normal distribution version. Good for noting how many standard deviations the data may be from the mean, and what possible outliers there may be.


Normal Variable

Standardised Variable

back to top

Link to census at school word doc on VIT's

!!!! NOTE !!!!

If you have manipulated your data (as above) and then want to run one of the iNZight VIT modules you must export your data, ie save it, then re upload it into the VIT.

Otherwise your changes will be lost.

This is straight forward to do, just remember where you saved it ready for the VIT

iNZight Inference & Bootstrapping

Most of this is covered in the AS3.10 Inference pages. Analysing a sample: Box plots, and sample statistics, bootstrapping, randomisation


iNZight Bivariate data - coming soon...


iNZight Time Series

Link to AS3.8 Time Series pages, or use the links on the images below:
Getting Started | Importing Data | Trend | Decomposing | Recomposing | Seasonal Effects | Predictions | Comparing multiple variables | Combining variables | Robustness | Summary


Linkk to Time Info Time Series Plot Decomposing Recompose Seasonal Effect Predictions Compare Variables Combine variables

back to top

Want to add any other ideas / suggestions on using iNZight,
please email


Convert to categorical Transform Variables Collapse levels Rename Levels ombine categorical variables Create New variables Reshape dataset Class intervals Rename variables Standardise variables Add to plot