Identifying Outliers in SPSS

[@tracithomas5427]

I have watched a few of your tutorials and am grateful for your thorough and easily understood directions and explanations. You have helped decreased my fear and increased my confidence!

[@annette2194]

You have single handedly saved my degree bless up Todd

[@MrsBobDylan]

This video was explained well and I think I understood the information. The step-by-step instructions throughout the video were helpful.

[@ms.brandihoward6871]

Thank you for your great explanation! It has made stats a little easier to answer.

[@ritusam1]

I think it is a great video. I did step by step as Dr. Grande instructed and I was able to get the outliers for my variables. Thank you.

[@remygooch1393]

Very helpful and informative video on identifying outliers, especially remembering to rerun Z scores after adjusting data.

[@ashleymegahan5966]

This video was very informative. I found it interesting to learn how to determine the outlier (-2.68 and 2.68). It has been a few years since I have used SPSS, but I think this video did an excellent job at showing how SPSS can organize your data

[@yardenlevy3182]

You are the best Dr. Grande

[@TheWolvesII]

Thank you for your videos! I'm wondering what to DO about outliers? What are the options, what is usually expected from students vs. mayeb researchers, what is most common...?

[@kzlorenz]

Hi Todd, nice video but slightly complicated method. Many thanks for your help.

[@DrGrande]

You're welcome - thanks for watching.

[@hafizahusairi]

Thank you! I more understand after watching your video =)

[@ushalincoln2546]

I think I remember some of this from undergrad and I think can be very useful when analyzing data.

[@ayrton56612]

When I run a box-plot I usually only pick out the extreme outliers with the little star sign. What would the z-score for an extreme outlier be?

[@ebrahimlari9241]

Thank you for the well-explained video. How about the outliers within the groups? What is the best test for those?

[@Jennalouise139]

Why any value less than -.2.65 or greater than 2.68?

[@journeymantraveller3338]

One definition of outliers is data that are more than 1.5 times the inter-quartile range before Q1 or after Q3. Since the quartiles for the standard normal distribution are +/-.67, the IQR = 1.34, hence 1.5 times 1.34 = 2.01, and outliers are less than -2.68 or greater than 2.68. Hence for normally distributed data, the probability of being an outlier is 2 times .0037 = .0074. This is less than 1%.

[@gerrygramozis544]

@@journeymantraveller3338 Do you have a reference for that definition of an outlier? Thank you.

[@ajboylan9338]

Where are you getting the numbers -2.68 and 2.68?????

[@amandagerrick5407]

I also found this video a little difficult to understand; however, I appreciated being walked step through step.

[@greadion4]

Hi Dr. Grande. I love your explanation, but I do not understand what you mean by using Z scores for population and box plots for the sample. Isn't the box plot representing the same data? The IQRs have not changed. I would appreciate a response. Thank you :)

[@darthnikki677]

This video has been really helpful, but how do I know if my distribution is normal to use the z-score of 2.68? Do I need to run a further test in SPSS?

[@zureezee1887]

This video is well explained, but for someone who is not a statistician it is difficult to understand.

[@riase]

Why are there these new variables z scores?

[@gabegransden]

What does SPSS mean? Statistical packages for the social sciences?

[@tarkgh4656]

sir how do you calculate the achievement gap of a group of students after and before an intervention?

[@dsavkay]

Great, thanks! 💯

[@kimeeshareedwalker3263]

The calculations are little confusingWorking the problem out step by step first will help understand outliers But this introduction was helpful.

[@diane3460]

I agree I liked the step by step, but still found it confusing.

[@lennartmenssink9637]

You have to put the option "Save standardised values as variables" otherwise it will not work!

[@harshabhogle1020]

How about skewed data

[@phongthanh8147]

🖒