Laura's Notes

….And stickers too!!

We would like to welcome Phil Assheton to the Library team and more specifically welcome him to MASH. Phil is our Maths specialist in MASH with a background in computing, maths and finance.  He’s gone part time in his other job, in financial modelling, to provide support at the centre on Thursdays and Fridays.  Many of you may recognise Phil as he’s been expertly volunteering with us for some time now.

Contact Phil directly Passheton@lincoln.ac.uk or via our general email MASH@lincoln.ac.uk.

Having Phil on board has enabled us to offer more drop-in hours. MASH drop-in can be found in the Learning Development room on the ground floor of the Library.

Our hours are now:

  

 

  Tuesday: 10-1

  Wednesday: 1- 4

  Thursday: 10-1

   Friday: 1- 4

 

Once again, this is something i’ve been asked a fair about recently…’How is Cronbach’s Alpha found using SPSS’

There is a good walk-though on page 196 of Hugh Coolicans book, Research Methods and Statistics in Psychology (5th Ed.). This book explains the use of this reliability test, it’s limitations and offers a guide to interpreting your output.

Here is a screen-by-screen walkthrough of the process:

Choose Analyse

→ Scale

→ Reliability Analysis

Move the variables you’d like to be considered from the left box to the items box on the left. Make sure the model Alpha is selected (it tends to be as default). Select Options  and choose what you would like to see in the output. I opted for Scale if item deleted, this is a preference of mine, but not necessarily what you’ll always want. I suggest to look in the Coolican (2009) book for assistance with your choice at this point.

 

Once you click Continue and then OK,  SPSS will produce an output similar to this:

The second table, the Reliability Statistics, gives you the Cronbach’s Alpha value.

Refer to Coolican (2009) or other statistics book (plenty available in the library) for further help on interpreting the results.

 

Reverse scoring is something that I’ve been asked about a good few times. This useful when you have collected data on a questionnaire with Likert scale answers (i.e. 1 to 5, 1 to 7, 1 to 10 etc.). Often you find that some questions in the set refer to a score of 1 being low and other questions 1 is a high score.  What you would need to do is reverse ONE set of questions, either the 1 = low set, or the 1=high set, so that all the questions are consistent.

I tend to suggested to students to work on the data in Excel before you transfer it into SPSS. I suggest this as students are often much more comfortable Excel than SPSS.  However, there is a way to do it in SPSS and here is a really good, clear video showing exactly how to do it: