Correlation analysis using Excel

Поділитися
Вставка
  • Опубліковано 1 гру 2024

КОМЕНТАРІ • 107

  • @Leon-om1lg
    @Leon-om1lg 6 років тому +34

    This video saved my life. I thought for sure that I won't complete my thesis. But because of this, I finished it! If it weren't for this vid, I wouldn't be able to have my thesis defense. Thank you so much!!!!!!!!

    • @ChristianHigherEd
      @ChristianHigherEd  6 років тому +3

      I'm very glad to hear this. Best wishes.

    • @thabethmatoda913
      @thabethmatoda913 2 роки тому

      Me too i was actually crying 🤣 😂🤣 😂

    • @thabethmatoda913
      @thabethmatoda913 2 роки тому

      I didnt know what to do...I just started crying because had no hope.Had only 3 days left before i submit my thesis.

  • @ijustfollowthesun
    @ijustfollowthesun Рік тому

    Thanks so much. You're still helping people 10 years later. THANK YOU!!!!!

  • @aldehoyos6803
    @aldehoyos6803 5 років тому +2

    Finally!!! Someone who can explain correlation without going all over the place!!!! thank you!

  • @PAYLAYP
    @PAYLAYP 8 років тому +12

    This video has just saved my ass more than you know in the dying days of writing my dissertation... thank you so much

  • @angelikakachnowicz6759
    @angelikakachnowicz6759 5 років тому +2

    I don't know how to thank you enough ! such an easy and straight forward guidance ! God Bless.

  • @zubaidahyusoff8611
    @zubaidahyusoff8611 10 років тому

    this is a very easy tutorial to understand for someone who does not have any SPSS background at all. THANK YOU!

  • @sadiakhan6227
    @sadiakhan6227 8 років тому +1

    Really appreciated. I just finished my 100 mks college project with the help of your tutorial. Great work. Thanks :)

  • @fsolanosmg
    @fsolanosmg 7 років тому +2

    Great thorough breakdown! Very beneficial as I approach writing my dissertation!

  • @MAli-nw7ix
    @MAli-nw7ix 7 років тому +2

    Very helpful video, Chris, thanks for the effort. Tremendously helped me in my postgrad dissertation. Cheers!

  • @DetHanob
    @DetHanob 11 років тому +1

    This is so useful and clearly explained. Thanks immensely.

  • @jessicapreston9808
    @jessicapreston9808 10 років тому +3

    This was incredibly helpful! Thank you! I have a horrible DS teacher at the moment..I wish he explained as clearly as you :)

  • @lifespotwithmaryanne
    @lifespotwithmaryanne 3 роки тому +1

    Wow this is helpful.I really needed this

  • @mahmoudassaad8212
    @mahmoudassaad8212 7 років тому +2

    Thank you for Sharing This Chris. Very clear

  • @johng5295
    @johng5295 4 роки тому +1

    Thanks in a million. Awesome. Very well explained. Where have you been all these years.A+

  • @michaeldever9055
    @michaeldever9055 7 років тому +1

    love your videos. Basically just saved me for exams,

  • @jinhailongwang
    @jinhailongwang 10 років тому

    awsome video, very descriptive. Well done Chris

  • @edyta3665
    @edyta3665 6 років тому +1

    Excellent tutorial. going to look at your other videos. thx

  • @georgeoguejiofor9684
    @georgeoguejiofor9684 10 років тому

    Some of my statistical doubts were cleared . It's impressive!

  • @uduakokpanefe3944
    @uduakokpanefe3944 3 роки тому +1

    Thank you Thank you Than you. Just what I need.

  • @yingzou5484
    @yingzou5484 9 років тому

    thank you so much!! I am doing my dissertation. This is help me a LOT!!!!!

  • @techexpedient
    @techexpedient 4 роки тому +1

    good video

  • @sopavanh
    @sopavanh 8 років тому +3

    Very helpful VDO! thank you man!

  • @RhodaYap
    @RhodaYap 9 років тому +2

    Super helpful and relevant; and easy to follow. Thank you and I've subscribed! :)

  • @babyslug
    @babyslug 9 років тому +1

    this was extremely helpful! Thank you so much! Definitely subscribing.

  • @Andreasofspades
    @Andreasofspades 11 років тому +1

    Amazing, thank you! Using this for my quantitative analysis for my dissertation!

  • @dr.denycedaniels9620
    @dr.denycedaniels9620 10 років тому +1

    Thanks!! I understand the concept, with your assistance.

  • @samuchayabehera842
    @samuchayabehera842 6 років тому

    really sir it is very usefull to me....

  • @saadrehman7769
    @saadrehman7769 7 років тому

    very well explained. thank you sir.

  • @RafaelMilaniMedeiros
    @RafaelMilaniMedeiros 6 років тому +2

    Sharp and clear! Thank you!

  • @elizabeththomas3582
    @elizabeththomas3582 3 роки тому

    This was soo helpful ! Thank you so so much!

  • @adeilajaggernauth8295
    @adeilajaggernauth8295 8 років тому +1

    Very clear and informative video. Can you tell me how to address the issue of missing data when running correlation. Do i code it or leave it blank when running the test?

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому

      There are a couple ways to address missing values when running a correlation in Excel: delete the cases missing data, delete the variables with missing values, or impute the values for the missing data. This website has some good suggestions and explanations: www.real-statistics.com/descriptive-statistics/missing-data/
      I hope this helps.

    • @adeilajaggernauth8295
      @adeilajaggernauth8295 8 років тому +1

      Thank you very much!

  • @simplysanyal1986
    @simplysanyal1986 11 років тому +1

    thank you very much.this is why youtube rocks.

  • @MandozaII
    @MandozaII 9 років тому +4

    Thank you very much! This helped a lot!

  • @MyChidananda
    @MyChidananda 6 років тому +2

    How can we use Pearson correlation in this case? Likert scale is an ordinal variable, we need to Spearman test.Correct if I am wrong

  • @pathoyoutubification
    @pathoyoutubification 10 років тому +1

    Awesome job...

  • @ravinadageri9540
    @ravinadageri9540 5 років тому +1

    its very sensible lesson

  • @melissahenehan1685
    @melissahenehan1685 7 років тому +1

    Very helpful thank you for sharing!

  • @anfalberier3501
    @anfalberier3501 4 роки тому

    @chris Olson if my survey questions have sub point, which data should I use for correlation analysis. For example questions 1 has a, b and c. Question 2 has a, b, c and d sub-questions. How could create the correlation analysis. Should I average each question then run the analysis

  • @VKjkd
    @VKjkd 8 років тому +1

    Very helpful. Question - If you have percentages (overall satisfaction) by group or brands can you correlate them or ate there major issues with using aggregated data in this way?

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому +1

      VKjkd I'm not sure I completely understand your question, but you may want to consider a Chi Square test ua-cam.com/video/UPawNLQOv-8/v-deo.html

  • @fernandojackson7207
    @fernandojackson7207 7 років тому +1

    Also, if two independent variables are highly correlated with each other, does it make sense to get rid of one of them ? I assume if the two are highly correlated they do not both contribute on different aspects of the dependent variable, i.e., either variable explains the same underlying aspects as the other?

  • @abdiladif625
    @abdiladif625 8 років тому +2

    @chris olson if i wanna to find out four variables to there correlation how to do , i tried but does not work

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому +2

      abdilatif mohamed jama You may want to run a partial correlation analysis on all four variables or a multiple regression. Here are links to some tutorials that may help with that. Good luck.
      Partial correlation: www.listendata.com/2013/02/excel-formula-partial-correlation-matrix.html
      Multiple regression: www.excel-easy.com/examples/regression.html

  • @b.b9774
    @b.b9774 7 років тому +2

    Thank you! Very helpful!

  • @kanchanasewwandi7579
    @kanchanasewwandi7579 5 років тому

    I use a likert type questionnaire (5 point scale) in order to measure six variables: the questionnaire I have three different sub scales and each one has likert items which measure one variable. I want to find the correlation between the variables. How to calculate it in Excel?

  • @eyelsi6109
    @eyelsi6109 4 роки тому

    Hi, Chris!
    I want to perform Concurrent Validity and Test-Retest Method.
    May I use the same set of scores I used in Concurrent for the Test-Retest Method?
    Thank you so much in advance!

    • @ChristianHigherEd
      @ChristianHigherEd  4 роки тому

      Hi Ey, I'm sorry, but I don't understand your question.

  • @MohamedGaber111
    @MohamedGaber111 4 роки тому

    Thanks.
    Could I use it for event correlation from event log?
    Is there a method to find out the frequent of an event from events log?

  • @nkeepfer
    @nkeepfer 8 років тому +1

    You say that you cant calculate the significance of the correlation, but within the "regression" option on the toolpak, you can see the p-value listed. I don't understand what you mean by this statement, please explain

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому +2

      Yes, the Excel regression analysis does produce a p-value, but the correlation test does not. That is one of the limitations of that analysis when using Excel.

  • @barristars5328
    @barristars5328 6 років тому +1

    GOSH!! you saved me there

  • @asw072
    @asw072 4 роки тому

    So sir.. In case of negative correlation coefficient.. Would the effect still be large?

  • @raghadali7998
    @raghadali7998 7 років тому

    Very helpful, Thank you !

  • @MohamedGaber111
    @MohamedGaber111 4 роки тому

    Could we do analysis between a several elements and build a heat map?

  • @teacuppy6610
    @teacuppy6610 6 років тому

    but this way it only shows the linear correlation through pearson coefficient doesn't it ? if you for example checked your scatter graph for this data (my situation with thesis) and you could see that lets say polynomic trendline is more accurate than linear one, is it still okay to use this way for calculating correlation ?

  • @cavejohnson3898
    @cavejohnson3898 10 років тому +1

    So good! Thank you!

  • @fernandojackson7207
    @fernandojackson7207 7 років тому +1

    Why can't we do a significance test in Excel? Is it not included in the Analysis Pak? Wouldn't we just do an Anova to test for significance and then do a pairwise test if we reject the initial that they are all zero?

    • @ChristianHigherEd
      @ChristianHigherEd  7 років тому

      Fernando Jackson I am not sure why Excel doesn't provide a significance test for the correlation analysis, but I found another video that describes a work around. Specifically, it suggests running a regression analysis. When you only have two variables in the regression, the Multiple R output is the same as the Pearson r and you can use the significance findings from the regression analysis. Here is a link to that video:
      ua-cam.com/video/vFcxExzLfZI/v-deo.html
      I hope it helps

    • @fernandojackson7207
      @fernandojackson7207 7 років тому

      Excellent, thanks for both replies.

  • @jeanettezarb2888
    @jeanettezarb2888 5 років тому

    Does this work only work on numerical data? Is it possible to do it with sentences from a survey?

  • @devangidesai84
    @devangidesai84 10 років тому

    Hello Chris ...I have a survey data...can I do the correlation analysis between Income and amount willing to spend on any product / service ..its for my hypothesis

    • @ChristianHigherEd
      @ChristianHigherEd  10 років тому +1

      Yes, if both variables are continuous. In other words both variables should be in individual "dollar" increments, not categorical groups such as $10,000 to $15,000, $15,001 to $20,000, etc.

    • @devangidesai84
      @devangidesai84 10 років тому

      Chris Olson Well thank you for your feedback. What if we the data in categorical groups ? Will the result be incorrect ?. Please let me know if the some other test that can be performed ...thank you

  • @tarikusnumidica8229
    @tarikusnumidica8229 7 років тому +1

    thank you chris its perfect

  • @mariamhusseina.6890
    @mariamhusseina.6890 8 років тому

    How to simulate the cross correlation in Matlab between two signals

  • @croaker260
    @croaker260 8 років тому +2

    WHat do you do if your two sets of data are NOT side by side?

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому

      Steve Cole There are probably more sophisticated ways to do this in Excel, but I would just paste the columns to be tested next to each other and run the analysis. I hope this helps.

    • @croaker260
      @croaker260 8 років тому +1

      That's why I ended up doing anyway. It seemed like the simplest solution, but in the future another assignment they may not be quite that easy. So I was hoping for a another solution. Thank you so much for responding. I'm in the healthcare budgeting class and you this video explained in very short time what my instructors lectures could not thank you

  • @adellaazer778
    @adellaazer778 9 років тому +6

    Can anyone recommend essay writing company?

    • @myrtiegrinder8687
      @myrtiegrinder8687 9 років тому +8

      *essay-writers.all-about.in* Finance paper was a total disaster for me before I found this website. I didn't know where to start, what to write, but they showed me everything and now I am sure that with their help I can cope with any possible assignment.

    • @wiseowlresearch675
      @wiseowlresearch675 6 років тому

      Wise Owl Research

  • @rodbrenbautista218
    @rodbrenbautista218 8 років тому

    how to get the pearson coefficient ? is that the average of all the coefficient ? thank u in advance

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому

      The Pearson coefficient is found in the correlation data analysis output. Its about midway through the video. Hope this helps.

  • @sampathjayasekara294
    @sampathjayasekara294 8 років тому

    Hello. How to conduct correlation and regression test for independent variables. Pls help

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому

      If I understand your question correctly, when you find a strong correlation between two variables in your regression model this is called multicolinarity. You should consider removing one of the variables from the model and run the regression again. I hope this helps.

  • @Drgeius2009
    @Drgeius2009 8 років тому +2

    thank you !

  • @harisankaruk
    @harisankaruk 9 років тому

    what is the point of measuring the mean and sd in 5 point scale ? mode makes more sense

    • @ChristianHigherEd
      @ChristianHigherEd  9 років тому

      Perhaps, but in my experience the mean is usually used when looking for a correlation.

  • @charlottegoode975
    @charlottegoode975 7 місяців тому

    Thank you so much

  • @cjs3298
    @cjs3298 5 років тому

    Thank you sir

  • @Chiu_Hindusthani
    @Chiu_Hindusthani 10 років тому

    thank you very much sir!

  • @MrsTealFish
    @MrsTealFish 10 років тому

    awesome

  • @cressnaa4183
    @cressnaa4183 8 років тому +1

    are this satisfaction data here measured in likert scale of 1-5

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому

      Yes

    • @cressnaa4183
      @cressnaa4183 8 років тому

      hello there thanks for the response. Is it possible to calculate corelation if the data are in frequency like for example
      price satisfaction
      1 28 36
      2 34 40
      3 26 22
      4 18 27
      5 10 16
      where 1= strongly disagree .... 5= strongly agree

    • @ChristianHigherEd
      @ChristianHigherEd  8 років тому +1

      No. If I understand your question correctly it looks like you have summary data and not raw data. Summary = X number of 1s, X number of twos, etc. for two subjects. You need raw data to run a correlation, I.e., one record for each response from each subject, as in the video.
      Hope this helps.

    • @cressnaa4183
      @cressnaa4183 8 років тому +1

      yes its a summary. great help sir thanks a lot. the headache is finally over

  • @mybabyeatsbroccoli
    @mybabyeatsbroccoli 10 років тому

    Thank you!

  • @BharatBhushan-sr4uy
    @BharatBhushan-sr4uy 7 років тому

    sir tel me how to analysis data on genetic diversity in rice using d squre statistics

    • @ChristianHigherEd
      @ChristianHigherEd  7 років тому

      Sorry, but I don't understand your question.

    • @BharatBhushan-sr4uy
      @BharatBhushan-sr4uy 7 років тому

      sir i want to know ,how we analysis data using mahalanobis d square statistics in excel

    • @ChristianHigherEd
      @ChristianHigherEd  7 років тому

      Sorry, I don't know how to do that.

  • @zaccblack103
    @zaccblack103 3 роки тому

    you said pearson correlation? aren't we supposed to use spearman correlation?

  • @anthonydilshanthamel6727
    @anthonydilshanthamel6727 9 років тому

    Thanks a lot

  • @ashishmore7242
    @ashishmore7242 6 років тому

    THNKUU SO MUCH

  • @E.G.21
    @E.G.21 27 днів тому

    👍

  • @ClayREZify
    @ClayREZify 9 років тому

    Subscribed! :)