95% of respondents will fall between the upper and lower limits (as determined by the regression using SPSS).

What is the lower limit? (To 2 decimal places)

Assume a value of '5' for the independent variable ('My sub (sandwich) was delicious'). What is the value of dependent variable (predicted) - 'I would recommend Subsay to my friends'?  (Use 2 decimal places).

Is the model significant?

What is the value of the co-efficient of determination on the regression between 'My sub (sandwich) was delicious' and 'I would recommend Subsay to my friends'? (Three decimal places)

Now run a regression using 'My sub (sandwich) was delicious' to predict 'I would recommend Subsay to my friends'. 

True or false.

The model using 'My sub (sandwich) was delicious is a better predictor of likelihood to recommend Subsay, than 'I felt satisfied after the sub' 

True or false.

This is a useful model, that shows that if we improve 'satisfaction' having had a sub (sandwich) we could significantly improve the likelihood of exisiting customers recommending Subsay to their friends.

Subsay knows that it can make the subs (sandwiches) bigger, so felt that they could address the issue of not  'feeling satisfied after the meal". Run a regression to determine how impactful this might be on the likelihood of a customer recommending Subsay. How good is the model. Put the value indicating how good the model is in the answer below. (Three decimal places).

Considering only the items 'I would recommend Subsay to my friends' and 'My sub (sandwich) was delicious' with a correlation coefficient r = 0.37 (p <0.001).

How best would you illustrate this data in a chart? Select one from the options below:

There is no indication that there is a linear association between 'feeling satisfied after the sub' (sandwich) and 'needing a snack to fill up'.

Considering only the correlations with the item 'I would recommend Subsay to my friends', which items did NOT have a significant linear correlation? Tick as many as apply.