Using SPSS. Using SPSS

Using SPSS SPSS is powerful software application for the quantitative statistical analysis of data. All of these practical sessions are assessed and c...
Author: Ann Whitehead
8 downloads 0 Views 2MB Size
Using SPSS SPSS is powerful software application for the quantitative statistical analysis of data. All of these practical sessions are assessed and contribute to 20% of the overall module grade. They are designed to provide you with experience of how data, collected as part of your final year project, can be organised, presented and analysed and expand on the theory given to you in lectures. It is important to attend them all in order to understand the assessed components. Each provides you with skills that are then built upon in the next practical. It is vital that you do them in order or else you’ll find that you can not do an activity as it was explained in the practical before! The activities are intended to allow you to put theory into practice. They progress from entering data into SPSS to more advanced statistical techniques such as difference testing and correlation. Using SPSS Log in to the PC and double click on the SPSS icon (go to start in windows and locate SPSS in program files and open the program…usually in the statistics tab). You should now see the window shown alongside. Notice the radio button (the blank circle) called ‘open an existing data source’ is checked (i.e., has a dot in it). Un-check this button by clicking in the ‘type in data’ button and then click OK.

Use this to type in new data

Use this to open a new data file

Use this to open one of the data files listed. So if you have saved work you will find it here. It is best to use a portable device as you may not be on the same pc every time and some students also have issues accessing their shared space on the University server.

1

You should now have a window open that looks like the one above – this is the SPSS data entry window and could either be open in ‘data view’ or ‘variable view’….see the two tabs at the bottom! One for data view and one for variable view. You can toggle between data view and variable view simply by clicking on the ‘data view’ and ‘variable view’ tabs. Make sure that the data view window is open at the moment. You enter numbers in on the data view page and enter the words of variable names on the variable view page! Ensure that you are familiar with this or you will be entering words into SPSS as data and SPSS does not like this and you will not be able to analyse your data!

The overall presentation of SPSS is much the same as Excel: information is entered into cells and may then be subjected to analysis. However, SPSS has many more statistical techniques available than Excel and hence why it is the most widely used computer software in sport and exercise departments.

The data to be entered for this practical is provided in Table 1 below. Table 1. Anthropometric and physiological characteristics for eight well-trained runners. Participant Training history Mass VO2max number (years) (kg) (L/min) 1 7 59 4.11 2 10 68 5.08 3 12 79 4.87 4 3 72 3.99 5 6 67 3.76 6 8 65 4.21 7 13 76 4.65 8 11 74 4.86 Note: VO2max = maximal oxygen uptake determined during incremental treadmill running.

2

Titles of variables go in the grey boxes ALWAYS and never in the white boxes! These spaces are for the numbers! NEVER type in words in these spaces!!! Remember SPSS deals with numbers not words!!

In cell 1 underneath column var00001 type the number 1 and enter the remaining data for Participant numbers (i.e., 1 to 8 below each other). The entries should appear as shown on the left. Continue in the same fashion entering VO2max data in the next column, mass and training history data in the same way in the next three columns. Having entered these data the first thing to do is give the file a name and save it. You can save onto data stick or onto your space on the university drive – make sure it is in your space as you may not be on the same computer next week and even if you were it may be deleted!

Question Why are the participants being referred to by number and not name? When you collect your own data for your project you will be the sole source of information that can link the data to a participant. In your data set you may have their name – BUT – the first column should still be the participant number. Remember – save your data often to avoid loss in case of unexpected crashes. We now need to define our variables (name them) so we know what the numbers are measures of. Now we need to click on the variable view tab…remember the data view page is just for numbers

3

and not letters!.

Your screen should now look like this:

Now we can begin to write in the names of our variables under the name column one under the other! So when we click on the data view again the names of our variables appear above the numbers in the grey boxes. Some SPSS programs only allow a certain number of letters to describe your variables. If this is the case you’ll have to shorten the name (e.g., partic)! Underneath the named participant variable write your next variable name (i.e., training history (hs)). You will quickly realise you cannot do this! You are likely to get the following message:

4

It is because you’ve used spaces and brackets which SPSS does not like! So try rewriting it using something like: traininghist. Then create a name for the variable of mass underneath and then VO2max under that. Your file should look something like:

Now it is important to remind ourselves what the variables mean because we have shortened many of them. So in the column LABEL write the full titles of the variable names, including spaces and their units of measurement in brackets. Your file should look something like:

Now we can switch back to the data view and your file should now show the variable names in the grey boxes across the top of the file and if you glide your mouse over these the full label names should flash up to remind you what these variables are and what their units of measurement are. See next page as an example: 5

Having created the data file we will now perform some basic operations on the data. Looking at some statistical data It is important to report descriptive statistics for all measured variables in the study. SPSS will do this for us very easily as follows. Click on Analyse, Descriptive Statistics, Explore and the following window should appear.

Click on each variable you are interested in and transfer each variable (i.e. training history, mass, VO2max) over to the dependent list section...click on the name and then the arrow opposite the dependent list box and it will move over. Do this one variable at a time by clicking on it, it will turn blue and then click the arrow to move it over. Then do the next and so on. Then click OK.

If things have worked well SPSS will take you to an output window, which should contain information as shown overleaf.

6

The most important information here is the following: Descriptives Training history (hours)

Mass (Kg)

VO2max (L/min)

Mean 95% Confidence Interval for Mean 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis Mean 95% Confidence Interval for Mean 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis Mean 95% Confidence Interval for Mean 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis

Lower Bound Upper Bound

Lower Bound Upper Bound

Lower Bound Upper Bound

Statistic 8.7500 5.9326

Std. Error 1.19149

11.5674 8.8333 9.0000 11.357 3.37004 3.00 13.00 10.00 5.50 -.459 -.578 70.0000 64.5636

.752 1.481 2.29907

75.4364 70.1111 70.0000 42.286 6.50275 59.00 79.00 20.00 10.00 -.328 -.396 4.4413 4.0364

.752 1.481 .17122

4.8461 4.4436 4.4300 .235 .48428 3.76 5.08 1.32 .85 -.058 -1.754

.752 1.481

The above output was obtained by the sequence of clicking on the output in SPSS, selecting Edit, Copy Objects – followed by pasting into this document in word. You should do this for appendicies.

7

However, you should NEVER copy and paste SPSS outputs into your work as results. You must extract the results you need from the output and create your own table! Note Tables need to be properly formatted as in the example shown on page 2. YOU WILL NOT BE CREDITED WITH ANY MARKS FOR SIMPLY CUTTING AND PASTING. Apart from the fact the formatting in SPSS is incorrect for practical reports, projects and research papers, the SPSS outputs will generally also contain information that does not need to be reported. Note:  Labelling our variables was worthwhile because SPSS has reported the data using the label, which makes it very clear as to what the statistics refer to.  We can see that no data was missed, since N (the number of values) is 8 in each case.  The minimum and maximum of each variable is given to that we can easily check the range.  Means and Standard Deviations are quoted to too many decimal places in some cases...in sport we generally report to 2 decimal places! It is unlikely that data will need to be reported to more than two decimal places. To report such data in the text of a report (i.e., in the results section) we can choose whether to use a Table or text. Using a Table will draw the attention of the reader to the data, which may or may not be an important consideration. You should not repeat results just to fill space! Useful Operations: CODING DATA/Grouping Variables Imagine that all odd numbered participants in the data set we are using are female and all even numbered participants are male. If we wanted to ask SPSS to compare data across genders we would need to code it in some way so that SPSS would know who was male and who was female. This is quite easily carried out as follows. Click on variable view, define a new variable called gender.

Click on the ‘values’ cell and a grey box should appear. Click on it and the following box should appear as overleaf.

8

In the value box put ‘1’ and in the value label box type ‘female’ then click add. Repeat this procedure for the males by adding ‘2’ in the values box and then type ‘male’ into the value label. Then click add. The new box should look like the one below.

When you have this, click OK and return to data view. Basically, you have just told SPSS that every time you type in the number 1 in the gender column you are telling it that that person is female. Remember, SPSS does not work with letters! Every time you type the number 2 in the gender column you are informing SPSS that that participant is male. Now it will be able to explore differences between these two independent groups. Now you can code for gender by typing a ‘1’ or ‘2’ in the appropriate cells in the data view (all odd numbered participants (females) receive a ‘1’ and all even numbered participants (males) receive a ‘2’.

Now we can ask for separate statistics on male and female participants from SPSS. To do this click on Analyse, Descriptive Statistics, Explore and you should end up with a window as shown overleaf.

9

If the variables you want to analyse are not in the dependent list transfer them over as you did before. Then click on ‘gender’ and transfer it to the ‘factor list’ – it goes in here as it is a FACTOR by which you wish to explore the data. This will be the same for any groups of independent nature you have (i.e., gender, sport played, level of performance, year of study etc). Now click OK.

You will get a lot of data in the output window. You can see that SPSS has now given you the descriptive data for both males and females and not just the whole group. You will get lots of descriptive data from this, from which you need to extract only what you want to report. In this case we are interested in the mean and standard deviation of the data for both male and female.

10

11