About | FAMED | Experiments | Screencasts | Blog | Photos | Contact | Home

General Statistics and SPSS
plus E-Prime, OpenSesame, DMDX and Superlab Screencasts

 mini demo

The source of Videoscreencasts was a project to create a series of screencasts to aid 1st year psychology undergraduate students learn how to use the statistical analysis package SPSS. At the end of the course it was decided to not only expand the content of these screencasts to cover other topics such as psychological experiment generators, but to also make the videos freely available to anyone who wants to learn about these topics. You won't find any screencasts about the more popular pieces of software (such as MS Word) due to the already large amount of material available on the web. Instead, the screencasts here are much more targeted towards those working in psychology. That said, many of the packages used will almost certainly be of use to academics in other fields.

To view the movies you'll need a copy of Apple Quicktime version 7. The movies are currently at different resolutions but will be standardized over time. Some videos are in 640x480, others in 960x600 or 1024x640 (really depends upon the reason they were created! Some targetted WebCT, others the web). New videos will be presented in 1280x800 and 960x600 versions.

Please click here to send an email with any comments you have.

Screencast List:

Links - other useful websites on SPSS and statistics:

SPSS

Introduction To SPSS

SPSS is a statisical analysis package. On the surface it looks rather like any spreadsheet out there but its functionality is quite different. This video gives you a brief introduction to the software.

Download Video (42.4MB) | Duration: 11:40

Creating Basic Tables

Basic tables can display a number of different summary statistics. This movie takes you through the steps of creating basic tables

Download Video (9.79MB) | Duration: 7:04
Speed Data Independent Groups.sav | Data used in the video

Bar Charts In SPSS

SPSS has a number of ways of creating bar charts. This video takes you through a one way through which you can create a bar chart.

Download Video (5.72MB) | Duration: 4:11
Speed Data Independent Groups.sav | Data used in the video

Boxplots

Boxplot are useful to get an impression of your data, in particular, outliers and the spread of datapoints.

Download Video (1280 x 800) (10.7MB)
Download Video (960 x 600) (9.4MB) |Duration: 5:36
Speed Data Independent Groups.sav | Data used in the video

The Explore Command

The Explore command is particularly useful when you have a factorial design and want to get some descriptive statistics about your data.

Download Video (9.6MB) | Duration: 1:42
Speed Data Independent Groups.sav | Data used in the video

Frequencies

Get the numbers behind a histogram.

Download Video (1280 x 800) (12.9MB) | Duration: 4:50
Download Video (960 x 600) (9.3MB) | Duration: 4:50
BMI.sav | Data used in the video

Histograms

Graphing frequency data.

Download Video (1.2MB) | Duration: 0:14
Speed Data Independent Groups.sav | Data used in the video

Correlations

Bi-viaraite Correlations test the relationship between two variables

Download Video (3.5MB) | Duration: 0:36
Hearing Loss.sav | Data used in the video

Scatterplots With Equation

SPSS allows you to create a scatterplot with an equation presented on the graph. This video show you how.

Download Video (2.7MB) | Duration: 0:40
Hearing Loss.sav | Data used in the video

Scatterplots Without An Equation

Here's how to create a scatterplot without an eqution on the regression line.

Download Video (5.0MB) | Duration: 1:21
Hearing Loss.sav | Data used in the video

1-Sample T-Test

A 1-sample t-test tells you whether the mean of a set of data is significantly different from a given value. For example, given the speeds of cars recorded on a motorway, can we say people drive significantly faster than the speed limit of 70mph?

Download Video (1280 x 800) (9.5MB) | Duration: 3:58
Download Video (960 x 600) (6.8MB) | Duration: 3:58
Speed Data 1-Sample.sav | Data used in the video

Independent Samples T-Test

Independent Samples and Paired Samples T-Tests tell you if the means of two groups are significantly different or not. T-Tests can be applied to data that is normally distributed and when the data in each group come from different participants or when each participant provides a data point to both groups. This video demonstrates how to perform the first of these comparisons - when each group is comprised of different participants, hence the name: Independent Samples T-Test.

Download Video (1280 x 800) (11.8MB) | Duration: 4:41
Download Video (960 x 600) (8.7MB) | Duration: 4:41
Speed Data Independent Groups.sav | Data used in the video

Paired Samples T-Test

This video shows you the last type of t-test, that for two groups were each participant contributes a data point to each group (hence the Paired Samples T-Test)

Download Video (1280 x 800) (9.7MB) | Duration: 3:50
Download Video (960 x 600) (7.1MB) | Duration: 3:50
Speed Data Dependent Groups.sav | Data used in the video

Mann-Whitney U Test

What about cases where we have two groups and we want to see if the two groups differ but the data is not normally distributed? In that case we can use one of two non-parametric tests. The first is called the Mann-Whitney U Test and test to see whether two groups are different when each group is made up of different people.

Note: If you are using version 18 of SPSS (also known as PASW 18) there are now two ways of conducting non-parametric tests. To follow the instructions shown in the video follow the same menu selection of Analyse -> Non-parametric Tests and then goto Legacy Dialogs. This will bring up the same menu that you see in the videp.

Download Video (1280 x 800) (13.3MB) | Duration: 5:05
Download Video (960 x 600) (9.7MB) | Duration: 5:05
Speed Data Independent Groups.sav | Data used in the video

Wilcoxon Test For Matched Pairs

The Wilcoxon test is the non-parametric version of the paired samples t-test - it is for testing whether two groups are different where each participant provides a data point to each of the groups. It differs from the t-test in that it is a non-parametric test; that is, it does not operate on the raw data like the t-test does. Instead, all data are first ranked and the ranks of the two groups are compared. It is used when the data are not normally distributed.

Note: If you are using version 18 of SPSS (also known as PASW 18) there are now two ways of conducting non-parametric tests. To follow the instructions shown in the video follow the same menu selection of Analyse -> Non-parametric Tests and then goto Legacy Dialogs. This will bring up the same menu that you see in the videp.

Download Video (1280x800) (15.9MB) | Duration: 3:33
Download Video (960x600) (10.1MB) | Duration: 3:33
Speed Data Dependent Groups.sav | Data used in the video

1-Way ANOVA (Between Subjects) Method #1

There are two ways of performing a 1-way, between subjects ANOVA is SPSS. This is one method.

Download Video (1280 x 800) (16.72MB) | Duration: 5:59
Download Video (960 x 600) (13.47MB) | Duration: 5:59
Speed Data 1-Way ANOVA.sav | Data used in the video

1-Way ANOVA (Between Subjects) Method #2

There are two ways of performing a 1-way, between subjects ANOVA is SPSS. This is the other method.

Download Video (1280 x 800) (11.90MB) | Duration: 3:43
Download Video (960 x 600) (8.76MB) | Duration: 3:43
Speed Data 1-Way ANOVA.sav | Data used in the video

1-way ANOVA (Within-subjects)

This video demonstrates how to perfom a 1-way, within-subjects (or repeated-measures) ANOVA in SPSS. The video uses imaginary data on drivers speeds recorded in the south of England, the north of England and Scotland. As well as the main ANOVA procedure, a Tukey's HSD is performed to examine where possible differences lie.

Download Video (2.33MB) | Duration: 0:57
Data unavailable

2-way ANOVA (Between-subjects)

This video is the first of 3 videos looking at the 2-way ANOVA. In this example, we'll look at how to perform a between-subjects analysis. The speed data this time has been organised into two groups (location - either south or north) and sex (male or female). A drivers speed is recorded in either the north or the south and thus each driver only contributes 1 data point (as they can't be both male and female!).

Download Video (10.7MB) | Duration: 3:31
Speed Data 2-Way ANOVA Between Subjects.sav
Data used in the video

2-way ANOVA (Mixed Design)

The procedure for performaning a 2-way mixed design (or split-plot) ANOVA is very similar to the 1-way within-subjects procedure. This time we just need to specify a between-subjects factor as well as a within-subjects factor. In this example, each driver has their speed recorded in both the north and the south (so location in the within-subjects factor) and the driver is either male or female (and clearly this is a between-subjects factor).

Download Video (10MB) | Duration: 3:24
Data unavailable

2-way ANOVA (Within-subjects)

Of all the ANOVA procedure this one tends to take the longest to get to grips with. It's not that doing the analysis is hard (it is essentailly the same as the 1-way within-subjects procedure). It's just the it is very easy to enter the factors incorrectly. In the demonstration, drivers speeds on both motorways and b-roads (1 factor - road type) and in both the north and the south (the second factor - location).

Download Video (20.7MB) | Duration: 11:08
Data unavailable

Chi-Square Test

The chi-square test looks for relationships in categorical data. Conducting a chi-square test in SPSS is not as straightforward as you might think. This video demonstrate why and how to go about performing the test in SPSS

Download Video (48.2MB) | Duration: 8:21
Download Excel Spreadsheet (0.01MB) | The Excel spreadsheet featured in the video.

Stats By Hand

These screencasts demonstrate how to calculate various statistics tests by hand. They were developed for the module PSY1205 Introduction to Statistics - a level 1 module in the School of Psychology at the University of Exeter. The course handbook was primarily written by others so I am unable to upload it here as present. However, each question is clearly visible in the screencast. Each screencast covers a particular question that students were set. Questions were devised by Prof. Andy Wills.

Chi-square Test

This video shows you how to calculate a Chi-square test when we have 1 variable (in this case, brand of ketchup).

Download Video (32.8MB) | Duration 12:38

This video shows you how to calculate a Chi-square test when we have 2 variables.

Download Video (41.6MB) | Duration 15:51

Z-test

These two videos show you how to calculate a z-score and check for significance.

Download Video (Problem Z1) (10.2MB) | Duration 3:28
Download Video (Problem Z2) (12.4MB) | Duration 4:09

Variance Test

These two videos show you how to calculate the variance test.

Download Video (Problem V1) (57.6MB) | Duration 10:33
Download Video (Problem V2) (36.9MB) | Duration 8:27

T-test (Paired-samples)

These two videos show you how to calculate a paired-samples t-test.

Download Video (Problem T1) (30.9MB) | Duration 10:31
Download Video (Problem T2) (34.9MB) | Duration 11:42

T-test (Unrelated-samples)

These two videos show you how to calculate an unrelated-samples t-test.

Download Video (Problem T3) (59.0MB) | Duration 19:14
Download Video (Problem T4) (71.0MB) | Duration 22:59

Correlation

This video shows you how to calculate a Pearson's Product Moment Correlation Coefficient.

Download Video (43.5MB) | Duration 15:40

This video shows you how to calculate a Spearman's Rank Correlation.

Download Video (44.2MB) | Duration 14:24

E-Prime

E-Prime is an industry standard experiment generator from PST. It takes a "drag-and-drop" approach to experiment creation but also allows for you to get your hands dirty by creating some custom code to get it to do exactly what you want. The interdace is very similar to Microsoft's Visual Basic 6.0 so programmers of that language should feel right at home. Everyone else - dont' panic! E-Prime is easy to get to grips with.

Creating A Simple Experiment In E-Prime

This demonstration is designed to get you started with E-Prime. The video starts off by describing a simple experiment and one possible way of implementing it (in any experiment generator). It then moves on to show how the experiment can e created in E-Prime. If you want to work through the example, you can download the cat face photos used in the video here.

Download Video (42.8MB) | Duration: 33:39

Analysing Data With E-DaiaAid

After you have run an experiment with E-Prime you will obviously want to analyse the data. E-Prime creates a custom format data file which can be read using the E-DataAid package (part of the E-Prime suite). E-DataAid allows you to extract data from the data file and export it to other packages such as Excel or SPSS. This video demonstrates how to use E-DataAid to analyse the data from the cat faces experiment, creating in the previous tutorial. Since E-Prime creates a seperate data file for each participant, you will end up with as many data files as people you have tested. Another program in the E-Prime suite, called E-Merge, can merge these data files together into a single file which E-DataAid can then analyse. The second part of the screencast shows you how to do this and finally, export data to SPSS for analysis.

Download Video (1280 x 800) (121.6MB) | Duration: 35:50
Download Video (960 x 600) (109.2MB) | Duration: 35:50

Opening More Than One Experiment At A Time (E-Prime V1.1/V1.2)

E-Prime version 1 does not let you open two experiments at the same time causing problems if you want to copy objects between experiments. This video demonstrates a trick to get two copies of E-Prime running concurrently to allow you do this this. This technique is not supported and I've found that exiting from E-Prime inevitably results in a crash (not before you've saved your experiment though). E-Prime version 2 does not require this trick as it will open multiple experiments.

Download Video (3.50MB) | Duration 1:28

Opensesame

Creating A Simple Experiment In OpenSesame

OpenSesame is a free, open source experiment builder based upon the Python programming language. If you know how to use E-Prime already, moving to OpenSesame is very easy. That said, it is very easy to learn, even if you have no experience of using a computer package to create experiments. The video uses the same cat faces experiment as the E-Prime tutorial but if you haven't viewed that video then don't worry - the video begins with an explanation of what the experiment is trying to achieve. You can download the cat face photos used here, and the completed experiment can be downloaded here. OpenSesame itself can be downloaded from http://www.cogsci.nl/software/opensesame.

Download Video (1280x800) (106.9MB) | Duration: 37:07
Download Video (960x600) (78.5MB) | Duration: 37:07

SuperLab Pro

SuperLab Pro is a psychological experiment generator from Cedrus software. The version used in these videos is an older version (2.0) but will give an introduction as to one way in which psychology experiments are built up.

Editing A SuperLab Experiment

This video takes a pre-existing experiment and modifies it. This task was part of the 1st year module Data Analysis II here at York. In the experiment, either the letter L, R, or F was presented on the screen. It could be forwards or mirror reversed and also it could be rotated through a number of angles. The participants task was to say whether the letter was forwards or mirror reversed. The students task was to modfiy this experiment so that only the letter F was used in 0 and 180 degree rotations.

Download Video (34.2MB) | Duration: 28:05

Creating A SuperLab Experiment

Most of the time an experiment is not prewritten and has to be created from scratch. This video takes you through the steps that are required to create a simple SuperLab experiment. This task was again taken from a pratical for the 1st year Data Analysis II module. The experiment to be created is a simple lexical decision task in which participants must decide if a string of letters presented on the screen form a word or a non-word.

Download Video (29.5MB) | Duration: 21:49

DMDX

DMDX is another psychology experiment package but this one has the advantage of being free. Unfortunately, unlike the others it has to undergo some configuration after installation.

Installing and configuring DMDX

This video demonstrates how to install and configure DMDX. It's not mentioned in the video but on most systems selecting "Select Video Mode" will automatically select the current resolution of the monitor. Thus, if you're using an LCD screen (and it's configured correctly) you can determine what your monitors native resolution is from here.

Apologies for the audio/video synchronisation problems that occur just after DMDX is installed. Somehow it gets back in sync not long afterwards so I decided to leave it as is.

Download Video (1280 x 800) (22.1MB) | Duration: 10:09
Download Video (960 x 600) (16.5MB) | Duration: 10:09
Link to the DMDX webpage

Snapz Pro X

Creative Commons License

SPSS and Statistics Screencasts by Chris Longmore is licensed under a Creative Commons Attribution-Non-Commercial-Share Alike 2.0 UK: England & Wales License.

Last modified: 23 October, 11:29 PM