SPSS for Beginners 2 – Frequency Counts and Descriptive Statistics

This video demonstrates how to calculate simple descriptive statistics in SPSS, such as frequency counts, means, and standard deviations.

0:00In this video, I’m going to show you how to do some simple analysis in SPSS.
0:05I already have some data popped in here, but I have explained what they are.
0:09Before I do, pause the video and enter these numbers into your own SPSS
0:13spreadsheet.
0:14These two variables represent gender and height. Variable 1 represents people’s
0:21gender information, and variable 2 represents people’s height information.
0:26Instead of having to remember this,
0:28let’s go into Variable View right now and change the names of these variables.
0:32Call variable 1 “Gender”. Call variable 2 “Height”
0:39The data for people’s heights are in inches.
0:43We’re getting numbers between 60 and 70 inches, which is about five to six
0:47feet.
0:48The data for people’s gender are not quite as clear. I’m not typing in “male”
0:55and “female”, even though i could, using a string variable.
0:58Because male and female are not numbers, and SPSS needs numbers to analyze,
1:04instead, I’m using a coding scheme of 1’s and 2’s where 1 represents
1:09female and 2 represents male.
1:12This is an arbitrary code. I could use 0’s and 1’s, or 5’s and 10’s or
1:17any two numbers at all. But 1’s and 2’s are simple, so iIm using those.
1:23Now a quick mention on the scales of measurements.
1:26Gender is an excellent example of the use of a nominal scale because the group
1:31the groups are qualitatively and not quantitatively different. So I can change
1:35my settings to reflect that.
1:37I’m going over to measure and change it from scale to nominal. Height, on the other
1:44hand, is a quantitative variable with fixed interval differences between
1:48scores and a meaningful 0, so we can leave this variable as it is set to
1:52scale, because it’s ratio. We’re almost set to start analyzing different
1:58aspects of these numbers, but there’s one last thing we can do for gender.
2:06It can be a little bit confusing to remember which one was male
2:11and which was female, 1’s & 2’s…I can’t quite remember…
2:15So we can assign value labels to those things. I go back to Variable View and
2:21under Values, I can tell SPSS what each value means. I type in that a value of
2:291 means female. It knows that 1’s are female.
2:35Same thing for 2’s and male. Pop those in there and hit OK.
2:41Now if i go back to Data View, I see all those 1’s &2’s have been replaced by
2:46females and males.
2:48If i want to turn that off and see the numbers again
2:51there’s a little button up here on the tool bar called value labels. Click that
2:56they’re back to numbers.
2:58Likewise, I can activate it or reactivate it under the view menu.
3:02I just check this box that says value labels and we’ll go back to the males and
3:08females.
3:09I think it’s more useful to leave it on, so I’m just going to do that. Now I think
3:15we’re ready to analyze these data. Probably one of the more simple things
3:19you can do is just count up how often things occur.
3:22It seems easy to do by hand for this data set, but remember if you’re dealing
3:26with larger datasets, counting things up by hand can get a little tedious.
3:31So let’s have SPSS do it for us. Up at the top
3:34go to Analyze. Analyze in SPSS is going to be your very close friend.
3:40Probably the menu you use most often and all the analysis options are in there.
3:44So under Descriptive Statistics, go to Frequencies
3:49Here we can see how often or how frequent different things occurred.
3:54This window pops up. You’re going to see this type of window a lot in SPSS.
4:00Basically, all the different variables we have are on the left and all the
4:07variables we want to analyze go on to the right. So you can decide which
4:11variables you want to analyze just by highlighting them and then moving them
4:15over with this little arrow box. You can move them back if you decide you don’t
4:19want to go so just kind of grab a hold of it and then drag it over
4:22whenever you want to analyze so decide whatever you want to analyze, gender is
4:26probably the best right now and when you’re ready click OK now your output
4:32window is going to pop up here
4:34I was saying this in the last video but one of the problems with SPSS is that it
4:39often gives you a lot more information than you want.
4:42so it’s a bit of a trick looking for the relevant information. Also the outputs
4:47aren’t exactly pretty
4:49so we need to kind of figure out what it’s showing us here
4:52and luckily this is a pretty simple table. It is telling us first off this
4:57little table, the n
4:58is saying we have 10 valid scores, none are missing
5:02basically we don’t have any empty cells in there. This table is probably the
5:06more telling one. It is telling us on the left are different groups, we
5:13have female and male, and in this frequency column
5:17how often are occurring. So we have 6 females, we have 4 males
5:21how often they are occurring. We have 10 miles total. Here’s percents which can convert to
5:25percentages: 60, 40, and 100%. Valid percent if we have a few missing cases
5:31it would it would divide it differently of the total number of available cells
5:35instead of all cells, and then cumulative percent which is saying sixty percent of
5:40people are female and a hundred percent are either female or male
5:45You know, we really don’t need those things. I think most the relative
5:49information is just frequency, right there. So we could do the same thing for
5:56height to create another frequency table for height, but that would tell us how
6:00many people are 60 inches tall
6:03how many people are 61 inches tall how many are 62 inches tall, etc. that’s
6:08probably not going to be very informative to us.
6:11I think a better way to convey the information, or convey just how
6:16tall people are, might be to calculate the average of their height.
6:20So, let’s try that.
6:21And just FYI, you can run analysis from the output window, the same menu is there.
6:26Just for simplicity or just for stability, I’m gonna go back to the data
6:31view of the spreadsheet. Go to Analyze and under Descriptive Stats go
6:37to Descriptives. We get the same basic window again. All the variables we have
6:44on the left, all the variables we want to analyze on the right, just pick height
6:48this time and move it over and before we hit OK i want to see what different
6:53options we have. Click on this button in the upper right.
6:56These are all the different types of descriptive stats it could calculate for us.
7:00Mean, standard deviation, minimum and maximum, those are pretty standard things. If we
7:04wanted, we could get the variance, and the range
7:07and the standard error of the mean. The sum is good too, I am going to click on SUM
7:11There are more options on here like what type of distribution we’re dealing with.
7:16Display order: if we had multiple variables we were analyzing it once and actually we can
7:21analyze as many variables we want the same time,but we only have one right now
7:26we can specify here
7:27what order they appear in. I really just mostly leave this in variable list.
7:32Hit continue and whenever you’re ready just click OK, that’s what’s gonna pop up
7:39slightly below our last table. So we see here is our variable height
7:46how many scores are in that variable, we had 10, the minimum score is 61,
7:51the shortest person or group. 80 was the highest score. 692 inches…that would be
7:57how long
7:58everyone would be if we kind of stack them up on top of each other, but i think
8:03this mean is the most important thing. The average height for our 10 people:
8:0769.2 inches with a standard deviation of about 6.27 inches, so what we
8:14can do now is we can start to break down height across categories. We can compare
8:18heights of different people from different genders.
8:23The easiest way to do that would be to go to Analyze, instead of
8:29descriptive stats, go to compare means. Pick that first option… means
8:36here we need to specify what are dependent variable is and our
8:40independent variable
8:42We didn’t really assign people to condition
8:48Gender would be what’s called a quasi-independent variable
8:51let’s pretend it’s the independent variable because we want to break down
8:55the different categories the different groups
8:57well that’s gender, and the dependent list of the things we want to measure, do
9:02calculations on, that would be heights, the inches. So click that and I think we
9:09should be good to go. We could see what’s under options
9:11A lot of stuff we don’t need. When you’re ready, just click OK
9:17
9:19Once again we’re getting more than we really need from SPSS, this table
9:23I don’t really see what it’s used for. The second table is I think the one we
9:28really want to see. This is where we see how tall
9:31females are compared to males we have females the average height about 68 . 3
9:35inches and there are six females contributing to that mean standard
9:40deviation about 6.9. Males about 70.5 inches for males. The standard
9:45deviation about 5.9, so it’s telling us that on average
9:49males are a little bit taller than females. We also have the total 69.2
9:55inches. That’s how tall people are on average. All 10 people.
9:59The standard deviations of about 6.27 that the same number we got up here from
10:04doing descriptive for all those people 69.2. So overall, frequency counts and
10:10descriptive stats or where a good way to take a peek at potential trends in your
10:14data and SPSS made it extremely easy to do so.

Related:

Leave a Reply