Some help in choosing a program and doing ANOVAs:

ANOVA: Doing the basics *illustrated using StatView, not SPSS, but principles are the same

ANOVA: Repeated measures

ANOVA: Doing RM in StatView

ANOVA: When not to use StatView

ANOVA: Doing it in SPSS

ANOVA: Power and effect size

An alternative to RM-ANOVA that addresses the sphericity problem, has more power, and lets you test alternative models: Colin's presentation on mixed-effects modeling  (in R, not SPSS),  and a data file from Baayen et al. to practice with (used on a Mac; PC users may want to first open in Excel). See also the tutorial by Quene and van den Bergh (2004), "On multi-level modeling of data from repeated measures designs: a tutorial", Speech Communication 43: 103-121 (online access is available through the UCLA Library)

About analyzing data from speech perception experiments:  
Analyzing identification data
Analyzing discrimination data (and comparing identification and discrimination)
Analyzing discrimination and identification data with D-prime (signal detection)
Analyzing ratings data
(in preparation:  Comparing multiple cues/cue weightings)

---

Excel (Microsoft): should be on every general-purpose computer in the lab.  You may be surprised at how much you can do just with Excel.
However, see this page by Hans Pottel about limitations of Excel for serious statistical analysis.  

• a Ling. 104/204 handout on how to do some basic statistics in Excel
• an excellent summary webpage about using Excel to make graphs
• FAQ: How to add error bars to bars in Excel charts?  Here's an answer

• Formerly the most popular Statistics Package used in the Lab; no longer sold and not the best choice to use, but seems to be installed on all our computers
• Inadequate for some ANOVAs with more than 2 levels of a factor, so use with caution.
  

• Allows Repeated Measures ANOVA with sphericity correction
• Fully supported by the statistics consultants on campus.
• The CLICC offers remote access to SPSS by their Software Shortcut website.

Go to SPSS Home Page

Go to UCLA ATS consulting page on SPSS, including their SPSS Starter Kit for new users
Go to ANOVA: Doing it in SPSS

Error bars in Excel (FAQ)

This question comes up all the time, and since other pages elsewhere with the answer come and go, we now include it here, incongruous as it may be.

The relevant Excel command is "Format Data Series", which can be reached either by clicking on data after the plot is made, or through the Format menu.  Under Format Data Series go to the Y Error Bars tab, and (generally) choose either 1 standard deviation, or the standard error.

Which one to use?  Depends what you want to show...

Standard Deviation: for a normal curve, +1 SD above the mean is about 34% of the total data under the curve; thus plus-minus 1 SD around a mean value in a graph covers about 2/3 of the total data.

Standard Error: like the SD of the distribution of the mean (square root of variance of the mean).  The standard error is proportional to the SD and inversely proportional to the size of the sample.  1.96 standard errors gives a 95% confidence interval for the mean.  That is, the 95% confidence interval around the mean is plus-minus 1.96 times the standard error (e.g. if the mean is 100 and its standard error is 1, then the 95% confidence interval around that mean is from 98.04 to 101.96).  It’s useful to show the 95% confidence interval on a graph because these intervals can be visually compared across means – if the confidence intervals don’t overlap then the means are reliably different.  But the standard error by itself (not 1.96 times it) is not so useful visually.  Similarly, the 99% confidence interval is the mean plus-minus 2.58 times the standard error.  (Note: these multipliers are sometimes adjusted up for small samples, to 2.13 and 2.95.)

 

last updated Dec. 2006 by P. Keating

 

back to the UCLA Phonetics Lab statistics page

Back to the UCLA Phonetics Lab overall facilities page