Quantative Research Guru only

profileMYangel10
AU_R7031_M1_SPSS.pdf

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 1 of 6

Hi, welcome to R7031 WebEx on Data Entry and SPSS, and preliminary working with descriptive statistics. My

name is Ann Weaver, Argosy University. This WebEx goes with Module 1 of R7031 Quantitative Statistics and

our goal today is to show you how to enter data in SPSS and do some beginning generation of illustrations of

data. What you see here is an SPSS spreadsheet file. SPSS is a dedicated statistical software package, in

other words it is designed to do statistical analysis. It differs from other spreadsheet software you may have

encountered like Excel which can do some statistics but is not a dedicated software for the working of

statistics. So in R7031 we’re going to use SPSS because it turns out to make data analysis a lot of fun.

Here is an empty spreadsheet, notice that you can just enter data in SPSS as such, 1, 2, 3, 4, 5, 6, you can

enter it by hand; but I find it more convenient to enter data in an Excel file and specifically the data that we

have for Module 1 in doc sharing and then transport it into SPSS because data manipulation in terms of data

entry in Excel is very straightforward and hard to mess up. So here is R7031 Module 1 data which you will be

using for the next couple of modules. Notice that the data are in columns and that the variable names are

across the top in the first row. I’m going to close this Excel file, go into SPSS and ask it to open that Excel file.

Now because I’m in SPSS the window that pops open for opening data says I’ll show you all the SPSS files,

but I’m instead going to down to the Excel files. There is R7031 Module 1 data and I’m going to open it and

you will see this window in SPSS and note here it says read variable names from the first row of data. Yes, do

that because I already have the variable names in there, I hit Okay and in short order SPSS opens up the

Excel file as an SPSS spreadsheet.

Now as in the Excel file the variables go across the top but they are no longer in an actual row of data. At the

bottom left hand corner of SPSS you have two tabs, one is Data View, the one we are on now, we are actually

literally viewing the data. The other tap is called Variable Fields, now the variable names that were across the

top row in Excel and in SPSS are now in the first column, so that’s the difference between Data View and

Variable Fields. The reason that we have this sheet here is that to a computer data are all numbers. So we

have to tell the computer what kind of data we have, in other words what the numbers represent. So this

corresponds to a couple of different things, let’s just start with each variable and take it in turn.

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 2 of 6

If you’ve had an opportunity to see the explanation of the variables online in Module 1 there is a table

explaining the variables something like this: we have 14 variables that we’re going to work with, now each of

the variables has an abbreviated name as well as a full explanation for it, in SPSS you can only use a variable

name that doesn’t have any spaces in it, and it should be short, so I would stick with 8 letters or less and I

wouldn’t add any numbers, SPSS can do some very strange things to those variables names with numbers in

them. So what we are going to do is we are going to tell the machine that the variable is symbolized and then

we’ll have an opportunity to put the information in more specifically.

So our variables are as follows: Attitude About Nature on a scale from 1 to 100; Trail, the minutes of working

out weekly; Runner, whether or not a participant is a runner; Yoga, the minutes of practicing yoga per week,

etc. So we are going to tell the machine first Nature Attitude, we can label it here, Attitude About Nature, and

these are data that were collected from people filling out a survey. So we’ll go right to the last column here and

it’s labeled Measure, but it stands for Measurement Scales. The Measurement Scales of data are as follows,

there are four kinds. There are categorical data, and there are numeric data. So the two categorical types of

data are called Nominal and Ordinal. Nominal Data are categories of an equal standing, no inherent order like

males and females are both a type of gender. Ordinal Data are categorical data that do have an inherent order

like freshmen, sophomore, junior, senior. Most surveys collect numbers on opinions but those data are treated

like interval and ratio data, the number data that correspond to the numbers basically you’ve been using your

whole life. SPSS refers to those ratio and interval data as scale data.

So here we tell the machine this code stands for attitude about nature and it is on a – it’s a scaled variable, in

other words ratio or interval. Trail is the next variable, it can be labeled as minutes working out weekly. Now if

you make a mistake in here you can go back in this cell and fix it as long as you haven’t gotten out of it. Once

you pop out of the cell to fix a mistake in here you have to type the whole thing over. Now Trail stands for

minutes of working out weekly on the trail, it also is a scaled type of data. In other words minutes, there can be

0, 1, 2, 3, 70 minutes, so we’ll call it a measurement scale of scale. It sounds funny.

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 3 of 6

The next one is Runner, this is whether or not a participant is a runner. So we can just label it runner, yes or

no. Now this is a different kind of variable because it is not scaled numbers anymore. In a survey somebody

would be asked to fill out well put a number 1 if you are not a runner, put a number 2 if you are a runner. So

we are going to tell that to SPSS in the column called Values. Here the value is the number that someone

would fill out in a survey, the label is what they mean. So here – I’m sorry, yes this is no, not a runner. If you

are a runner put down 2, yes am a runner, something like this. So what you do is you put the number that it

corresponds to the spreadsheet and then the label for it, and then you hit Okay and then they will be visible

here and later on when we use these data analyze those levels will be identified.

Now over here, the last column, Measure stands for measurement scales. We either have categorical or

numeric data, Runner is a nominal variable. In other words you have two options, yes you are a runner, no you

are not, either or. You know either are or not, this is a dichotomous variable, a nominal variable with two

categories. Yoga is minutes of yoga per week. That’s minutes, that’s a ratio or interval type data, we call it

scale in SPSS and we don’t have to label anything in terms of values because this is literally the minutes of

working out, of working with yoga per week. The next one, college year, is the year in college. Now this is a

categorical data and a variable, in other words freshman, sophomore, junior or senior. They have inherent

order, so we call them ordinal data, they have order and we can tell the machine what that, those symbols are.

So on the survey we said if you are a freshman fill in number 1, if you are a sophomore fill in number 2, if you

are junior fill in number 3 and if you are senior fill in number 4. So now we have some words that go with the

type of data, ordinal data, we have values and labels.

The next variable here is Class. This is a physical fitness class, in other words are you a superior athlete, a

champion, a super cat or an ultra cat. Now these are ordinal categories because a superior athlete isn’t as

good as an ultra cat in terms of physical fitness, but notice here instead of under type which have all said

numeric so far, this says strength. If you click on one of these cells in the variable view of SPSS you get this

little box in the corner, click on it, it gives you some more options.

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 4 of 6

Here the String button or the Radio button is highlighted. String stands for words. I added a string variable in

here to show you that you can use both words and numbers in SPSS depending on the nature of your

dissertation. If I leave Variable View for a minute and go back to Data View this is what I mean by a string

variable. You can add the word here. Now for example in college years we used a number that represented a

word, in this case 3 represented a junior. Now we went ahead and put in the words, and this option is available

to you depending on the kind of dissertation that you have. But because we have a string variable these are

not available to us, although we have told the computer these are ordinal data.

The next one is Bench, and this is the highest weight bench pressed by a participant in the study, that’s the

numbers 1, 2, 3, 4, 5 pounds, so its scale data is appropriate. Squat is the number of consecutive squats a

respondent reported that they can do. This is also scale data, so we don’t have to label anything else. 40 Yard

Dash is seconds of time to run 40 yards, that’s also a scale type data. PTF is total fitness points, also scale

data.

Now I’ve got two variables here, Gen Num and Gen Word, they both stand for gender and it’s just to show you

a different way that you can use numbers or words depending on what’s convenient for your data analysis.

Gen Num stands for gender by number, and in this case gender is a nominal variable, in other words these

are categories in which everyone has equal standing. Males are comparably a gender as females, so it’s

nominal data. And now we can fill in the values of it. In this survey a person who was a male was asked to put

in 1, and a person who was a female was asked to put in 2. Okay, there you go. So we have gender by

number and we supply the explanation of the numbers under value labels. Here we have gender in words. So

the type of data here is string and I’ll show you in Data View what the numbers look like or what the data look

like under Gen Word. If you go over here, see the words literally appear. So here you could have females put

in number 2 or you can type in the word female.

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 5 of 6

If we flip back to Variable View we just have two more variables to take care of this. This is Before Treatment

and After Treatment, this was a survey of optimism about fitness. So respondents filled out this survey, it was

– because it was survey data we consider it scale data and so it’s numeric type data. And then their optimism

after – let’s fix this. I left the cell, I’m going to have to redo it again. Optimism Before Treatment, Optimism,

and the treatment in this case is working out in the physical fun nature trail as per the research problem in

Module 1.

All right these are still scale data, so it looks like we’ve got everybody organized and defined as we should.

Now you can generate information in SPSS from the dropdown menus up here whether you are in the Data

View as we are now or in the Variable View, but I like to look at my data when I am working with them. Now

the other thing I notice here is for some reason you see that the numbers down here are vague, the numbers

here are highlighted and there is dots here. Just to be on the safe side, I’m going to highlight that and have

SPSS clear those cells because it might think otherwise we are missing some data. Why it should I don’t

know. But we have some then descriptive statistics that we’re introducing in this Module and basic SPSS data

handling skills.

So let’s take a couple of straightforward categories of data and see how to generate some graphics in SPSS.

Let’s take Runner. We have two conditions or levels of the variable Runner, one means no, I’m not a runner, 2

means yes, I am a runner. To look at that pictorially we can go from Analyze Descriptive Statistics to

Frequencies. Now this box, Variables, is empty but this box has all the variables that we defined. So we can

go to Runner here. Now notice that Runner has a different symbol with it than some of the other variables.

This triple circle represents nominal data. When I highlight this variable this button becomes active, I hit it and

it moves the variable over. In SPSS pictures are called charts and because we have nominal data we can ask

to generate a pie chart of this variable. I hit Chart, Pie Chart, Continue, Okay.

Now what will happen in SPSS is it will generate a second sheet, this is called Output and this is where the

tables and figures that you asked SPSS to generate will appear. So here is a Pie Chart of Runners, yes or no.

Methods and Analysis of Quantitative Research

©2012 Argosy University Page 6 of 6

Blue stands for no, not a runner, green stands for yes, a runner. So this quickly would be a way to indicate to

people that the sample is about evenly divided between people who are runners and people who are not. But

SPSS will often also generate a table of descriptive statistics for you, so here they have the name of the

variable, at the top the two levels, no, not a runner, yes, am a runner. The frequency, literally the number out

of the 30 people who participated in the survey who said they were not a runner, 16, that constitutes 53.3%,

about half of the people in our survey. 14 of the 30 people said yes, I am a runner, that’s 46.7% and these

percents we are getting by divide, were gotten by dividing the frequency by the total. So 16 divided by 30 is

53.3, 14 divided by 30 is 46.7. So in the dissertation you could use this table of descriptive statistics, you could

use the figure.

Now we have a couple of other nominal variables so we can do the same steps again. I can go to Analyze,

Descriptive Statistics, Frequencies and I can take Runner out by highlighting it and pushing the button that

way, but I can also holding the Shift or Control button down as I need to take some of the other nominal

variables, make sure it’s this Pie Chart there, Continue, Okay, and SPSS will generate not only all the

Frequency tables that you need but it also generates the Pie Chart. So here is an example of the Pie Chart

that you see in Module 1, you are in college you can see the division or the distribution of years in college of

the people participating in our study. Here is the distribution of people in the physical fitness classes and here

is a distribution of the sample by gender, half male, half female.