**News Alert:**Experts Exchange Confirmed as Safe in Cloudbleed Leak Read More

Solved

Posted on 2013-06-29

Hello Experts:

I have fundamental statistical background... currently learning my ways in SPSS.

Please see attached PDF. Contains some statistics (i.e., running multiple regression) and graphs.

I'd like to get some assistance in interpreting the data. From my perspective, the regression line indicates a "positive correlation" between the independent variables (I included 4 of them) and the single dependent variable. In fact, the R^2 = 0.908... which I think is considered high/strong.

However, looking at the ANOVA and Coefficients summary table, the p-value (I think that's "Sig.") is NOT significant... that is, it is shown as 0.524. From what I know, a significant p-value should be less than 0.05 or 0.01.

So, if I interpret the graph with regression line correctly, I would have expected the p-value to approach 0. Again, that's not the case.

Am I mixing things up? Given the limited amount of info included in the PDF, could someone please help better understand the data/information?

Thank you in advance,

EEH

Correlation----Obtain-Better-Und.pdf

I have fundamental statistical background... currently learning my ways in SPSS.

Please see attached PDF. Contains some statistics (i.e., running multiple regression) and graphs.

I'd like to get some assistance in interpreting the data. From my perspective, the regression line indicates a "positive correlation" between the independent variables (I included 4 of them) and the single dependent variable. In fact, the R^2 = 0.908... which I think is considered high/strong.

However, looking at the ANOVA and Coefficients summary table, the p-value (I think that's "Sig.") is NOT significant... that is, it is shown as 0.524. From what I know, a significant p-value should be less than 0.05 or 0.01.

So, if I interpret the graph with regression line correctly, I would have expected the p-value to approach 0. Again, that's not the case.

Am I mixing things up? Given the limited amount of info included in the PDF, could someone please help better understand the data/information?

Thank you in advance,

EEH

Correlation----Obtain-Better-Und.pdf

2 Comments

p is a measure of how likely it is that random data would produce what you see.

The reason your R^2 is high is because (as you can see) the line does really seem to fit well.

The reason your p isn't very low is because you only have 9 data points. The odds that random chance will look correlated is very high with such a small sample size. If you had a lot more data and the regression line still fit that well, you would see p be much smaller.

Great... that explanation is exactly what I was looking for. The information is based on a survey pilot. The actual survey will probably result in several hundred data points.

At this time, I'm merely investigating what analysis options are out there... basically preparing the upcoming data analysis phase.

Again, thanks for the valuable feedback.

EEH

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Title | # Comments | Views | Activity |
---|---|---|---|

formula how to get the number incrementor? | 3 | 23 | |

formatting - number format | 2 | 14 | |

Time difference between dates without weekend | 16 | 30 | |

VBA Code to loop through a csv file and select only the cells in a column that have names in them. | 8 | 16 |

Join the community of 500,000 technology professionals and ask your questions.