### Present Remotely

Send the link below via email or IM

• Invited audience members will follow you as you navigate and present
• People invited to a presentation do not need a Prezi account
• This link expires 10 minutes after you close the presentation

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

You can change this under Settings & Account at any time.

# Point-Biserial Correlation Coefficient

No description
by

on 26 April 2011

Report abuse

#### Transcript of Point-Biserial Correlation Coefficient

Point-biserial correlation coefficient This type of correlation is looking at two different types of data. Dichotomous Variable "Normal" Types of Data Naturally Dichotomous Artifically Dichotomous All this means is that the data is not numerical like "yes" or "No", "Male" or "Female", "Sith" or "Jedi" So for instance all Males in my sample will be given a 1 and Females will be given a 2 in my "study". This is data that we are all used to seeing in the forms of Ratio, Interval, Ordinal etc. Think Nominal Ordinal Ratio or Interval? For my next part of the study, I.Q. will be used. When Dichotomous data and "Regular" data fall in love they produce Point-biserial correlation coefficient. Dichotomous Data Interval Data Male=1 / Female=2 I.Q.= Well it equals I.Q. X=1 X=2 100
115
109
108
116
103
98 107
106
116
118
112
105
100 Gender: I.Q.: So after the Sorcery Called Stats we have produced three important calculations: r pb t df = = = +0.17 +0.59 12 Weak Correlation Critical Value of t is 2.18 Null hypothesis Excepted (Based on t-distributions table) * N=14:
n1=7 Males / n2=7 Females Hypothesis: Males will perform better then Females on standardized I.Q. tests. Results: We found no significant difference between the I.Q. scores of Females compared to Males (t=0.59,p>0.05,df=12). ~FIN~
Full transcript