Chi-square test in SPSS + interpretation

[@savannaleehodgkinson189]

This person taught this topic in 5 minutes and I fully understand, versus my 3 hour lecture by a Professor who just left me hella confused! Thank you!

[@bipukarki8117]

Yeah facts

[@giftypokuaaagyeman8108]

Fact

[@6eyeshollow_technique]

Lol... Same here but from different place

[@lukeliu7042]

I felt the same with my lecturer

[@saadamjad5224]

HAHAHAHA Samee heree broo , have a Final Exam tomorrow and was confused in the lectures but this is just an awesome explanation!

[@Numbaofthechurch666]

Cannot thank you enough, not only for the incredibly informative & accessible information but for having a voice infinitely nicer & more comprehensible than the other videos available!

[@ld1667]

so true

[@karyngariepy6820]

This video was by the most helpful one that I've watched. Literally have not understood this concept the whole semester until now and it was so easy. Thank you for posting!

[@ShesSoCoolLike]

This might help save me from failing my stats module! Very calming which is great for my stats anxiety too 😆

[@sherlockboggart685]

This is the first time I have seen a chi square test done in person. This helps me understand the material better.

[@adamuppsala1931]

Omg, I can't believe this. So I was looking for a tutorial about it and found this one done by Brunel University London, a place where I work. I even know the ASK office!!! Small world!!

[@e.hadzhimincheva5405]

You explain in such a pleasant and precise way! Lovely.

[@dr.chandankumar7798]

COVID 19 FORECASTING VIDEOS : 1. ANALYSIS AND FORECASTING OF COVID-19 DATA (WHOLE WORLD) WITH SPSS kzbin.info/www/bejne/haang5xomch8qc0&ab_channel=VIDEOHUBB 2. ANALYSIS AND FORECASTING OF COVID-19 DATA ( INDIA ) WITH SPSS kzbin.info/www/bejne/oF6TmqeagZWCe7s&ab_channel=VIDEOHUBB

[@konstantinosmarkourakis890]

saved my university degree!! love u!

[@mujahidali3686]

Explanation was clear and example was also relavant

[@orendainlicon]

I struggled so much and you helped me like you have no idea.

[@jeanvr777]

Thank you for this video. Straight to the point and easy to understand.

[@bjarbj944]

You are a life saver. One thing I would have liked to know is how do you explain what the Chi square number actually means? We are always told not to just consider p values alone.

[@katzistan]

This is so helpful, thank you for making this.

[@aa2g]

where is the other video for when the Chi-Square assumption is violated? Please send a link. What should I do then?

[@Jesuslovesyouinfinitely]

kzbin.info/www/bejne/qmipfqKkrriders

[@akpofureenughwure6177]

You can find the part two of this video by forwarding the link... kzbin.info/www/bejne/qmipfqKkrriders

[@rachaelharwood9063]

Thanks so much for saving my SPSS sanity. Makes perfect sense

[@tharii314]

You saved me from my assignment! ❤

[@fmomeni6271]

thank you excellent video, quick question, at 2:53 you explained that the value needs to be below 20%, if it is more we should use a different approach, what does that mean, I am working on a project where my value is 83% in my chi square test,

[@martinzezulka8418]

Thank for video, but I am little bit insecure in 3:57. Is really 0,035 less then 0,005 and hereby, the result is H1? Sorry for stupid question...

[@jrivers8074]

Great video, thank you for clarifying chi-square tests. One question: at around 4:10 you stated that we would "accept the alternative hypothesis." I thought we never accept any hypothesis, we either reject or do not reject the null?

[@njhmd8783]

it means the same. p value 0.035 is less than the alpha value 5% (0.05), thus, we reject the null hypothesis = we accept the alternative hypothesis. there is a statistically significant association between gender and smoking status.

[@christopherjamescj1331]

What if if we already have observed value cross tab , how to calculate expected values . If we dont have individual data just total summary count of observed values

[@starry.nights9740]

Thank you so much for this straightforward explanation

[@mpraveeng1972]

Finally, now I understood chi square. Thank you

[@DrAminaHussayn]

Precise and clear EXPLANATION ..... THANKYOU

[@ninagray961]

Definitely a great video, short but detailed and I actually get it now. Thank you!!

[@ernestosalazar5094]

Thank you so much. Really appreciate your video. Clear and well explained.

[@actuarykiller]

Nice video. I'm new to SPSS. Can you elaborate on what alternative approaches one can take if more than 20% have expected counts less than 5, please. Thanks!

[@moplikjh]

Fisher Exact Test

[@basmaakroutbrizard8018]

Hello, Could you please write down the name of the book you are referring to? Many thanks!

[@riase]

SPSS Survival Manual: A Step by Step Guide to Data Analysis Using IBM SPSS by Julie Pallant

[@lolhahalolhaha111]

You are a saint for us students

[@Ijazalijin]

thanks a lot very informative video i learn completely how to perform a chi square test in spss thanks one again

[@alessandragirlando9946]

Hi, which is the alpha value which is mentioned in 3:51 please?

[@CPTBacardi]

What if I told you that my Pearson r test shows a negative not significant correlation, but my Chi Square test tells me otherwise?

[@TheDriverpreneur]

What if the dependent variable is more than 2?

[@TheDriverpreneur]

My exact dilemma

[@63navygirl]

First off thank you!! I'm currently taking a Statistics in Social Science Course and I was so lost.

[@sereneSilvana]

Very useful, very well explained, thank you! But I do need to know what to do when the is less than 5. Where can I find explanations about the interpretation of the test in that case? Also, I encountered another problem. I am trying to determine if two independent groups (one studying painting, one studying music) is significantly different in terms of temperament. This made things more complicated I guess, because now one of my nominal variables has 4 values and the output of SPSS looks different - for example, I can no longer see the row of Fisher's exact test. Could you help?

[@bchen4593]

What a video! Really appreciate it

[@MrFromsomewhereelse]

Hi. I'm having difficulty with "expected counts". What to do if there are more than 20% of the expected counts are less than 5? In my case: 15 cells (53.6%) have expected count less than 5. The minimum expected count is .07. Thank you.

[@sodaleo1980]

+Cavid Seyidov Same problem here

[@sodaleo1980]

+Cavid Seyidov Got it. In case of small samples, Qui square is not good, and you must use Fisher's Exact Test. For our luck, this result is in the Chi-square's same table.

[@렘모르]

Hello, I hope someone can answer me. I just would like to ask when you already established that gender and smoker status have a correlation, how can you establish whether what gender smokes more than the other? Or which gender is more likely to smoke? I am having the same trouble with my own stats and I need to determine this. Thank you for anyone who can answer me.

[@m.hasnatakhtar187]

did you find the answer, im unable to understand that as well. can you help me

[@frndcmc22]

Me too

[@fezalikhan]

all was ok, but i did not understand your last point, about Phi value? i calculated mine, in different study, it was 0.24, so is it in my study, X has moderate effect on Y, or what table you consulted for level of association of effect for two?

[@AwokeDamtie]

Thank you very much for the appropriate information to give me!

[@kodyioia85]

So say I find that the Chi-square for one variable is significant, but the Chi-square for the other variable isn't - does this suggest a one-way dependency between the two variables? I.e. Variable "A" is dependent on variable B, but not the other way around?

[@wanhanissuffia9899]

May I know what is the difference between Chi Square Test and Spearman Correlation Test?

[@Dr.Dave718]

thank you so much for your video, in your explanation it was very precise and very CLEAR...... helped me out a lot

[@ruchiarya3163]

what we do if that a. point value is more than 20%, please do suggest what course of action should be taken

[@regabrielexv]

thank you! Very clear and complete!

[@creetybaral5137]

Thank you sooo much ❤️

[@alighorbi1837]

Thanks for your great explanation 👍

[@frndcmc22]

What if I want to see which geneder I mean male or female is significantly associated then what test should I do? I mean which test is equivalent to post hoq test for categorical variable?

[@aymansgarage]

How do you calculate the number of pairs tied on x and tied on y for these variables.

[@IR240474]

Thank you very much!! I have an exam in the morning!! hehe....

[@Faith_Ndunge]

Thankyou is not enough. i am grateful.

[@Be_that_person]

how come there is a significant relationship? Based on the result p-value is greater than .005 that means there is no difference? please explain it to me... I am confused

[@fcsmovich]

I'd ask the same question... I suppose the point is the author used not a collected data, but she used a self-made database. And she was keeping in mind an idea, that ordinarily men are seemed to be smokers and women to be non-smokers. And she made the database self to prove this correlation. But something went wrong....

[@katibsareeh9056]

Hi. Thanks for your great effort. There are 3 values in SPSS output tables regarding crosstab procedure, first is Pearson Chi square value, second is likelihood ratio value, and third is Fisher's exact test value. Which value to choose and report? I noticed that Pearson Chi square value, and likelihood ratio value are always shown in any SPSS crosstab procedure regardless the number of rows and columns and regardless the violation or no violation of Chi square assumptions, whereas Fisher's exact test value is only shown when there is more than 2 by 2 table and there is a violation of Chi square assumptions? Is my notice true or there is another rule behind that?

[@michelemcallister6673]

This really helped!! Thank you!!

[@FreakbeatJaime]

is child/adult a nominal variable? What about correct/partially correct/incorrect? Thank you!

[@salimalbitar]

well, child and adult are mostly nominal (depends on the coding and context) but imo correct/incorrect are scale

[@sunnys1n9h]

How do u input data into SPSS when you have observed results only, and the data you are given is a mean split for low/high social aggression and low/high for physical aggression with question answered using a 0-5 Likert scale.

[@Itania333]

You absolute hero. Thank you so much

[@EmmnauelPeace]

Good work, very informative and positive

[@IamVishalVS]

Is it possible to enter data of more than one option in a single cell and analyze in SPSS. (Example: value 1, 4 in a single cell)

[@V3erle]

I'm a little confused. When you get a significance of less than 0.05 you are supposed accept the H0 and not reject it, right?

[@emilytreen6270]

+Veerle Hendriks You reject the null hypothesis if you find significance below .05.

[@alisalm5022]

when we use 2x2 table , do we need to use the continutiy correction to calculate the finl p value ? i mean the opposite no. to the continuity correction should be the p value ? or im wrong kind regards

[@cindyg8571]

Thank you thank you thank you so much! This really helped me out :)

[@atikhomthienthong416]

My table is 2 by 3. Which one should I use: phi or cramer's?

[@AbdulHadi-us5yb]

Hello , I have a question please ! If we have primary data, and collected data with some 400 questioners. So do we need to split these questions variables vise. Like some questions for independent variable and some for dependent variable. Are we can run the chi square test on the overall result of the questions. I have a single excel sheet of all my questions with 1 and 2 value of results.

[@jaystevens9993]

Hi, what test should I choose if I had 3 categorical variables? Let's say Male-Female vs. Heavy Smokers, Light Smokers, and Non-Smokers?

[@chemstid30]

Hi Brunel, How could i analyse my result in case i have this note: " 3 cells (75.0%) have expected count less than 5. The minimum expected count is ,60." And what does it mean ? Best regards

[@wmckin11]

chems tid The Chi Square test assumes that each cell has an expected frequency of 5 or more. As this appears to be the case with your results, you should use Fisher's Exact test - it has no such assumption. Fisher's should be used in cases such as yours. I fear this answer may come a bit late for you but hopefully will help others!

[@Swankyvali]

Urgent question: If the chisquare p value is less than 0.05. Does this also mean that there is a statistically significant differences between male and female in whether they smoke or not?

[@ycdantywong]

Is there a video showing what to do when more than 20% of the cells have expected counts less than 5? Much appreciated this is very helpful !

[@sodaleo1980]

+Yiu Chung Wong In case of small samples, Qui square is not good, and you must use Fisher's Exact Test. For our luck, this result is in the Chi-square's same table.

You are right but Fisher's Exact Test can only be performed if we have a 2x2 contingency table. For example we have gender and smokers-nonsmokers like in the video. male-femal vs. smoker-nonsmoker is a 2x2 table so we can use also Fisher's Exact Test even if proportion exceed 20% but if we have gender and education with 3 options(like undergraduate, graduated, master student) that will be a 2x3 table and that time if proportion exceed 20% we cannot perform Fisher's Exact Test.

[@voldemort50]

@ Thank you for the explanation. In case the proportion exceeds 20% and we have 2x3 table, which test should we use?

[@draculu]

So the exact significance (2-sided) doesn't tell us anything about the relationship, it doesn't matter when testing significance?

[@manthanjoshi9144]

What to do if more than 20% cells have expected count less than 5?

[@allenhenderson8402]

What about if measure of my scale against only one demographic and their scores?

[@paulfiskmofolo4910]

If someone tries to determine if socioeconomic factor's were significant different between thr two set of groups e.g gill net and seine net. How can we go about it?

[@bharatmirchi]

What stats to use if expected count has been violated.

[@farfurcia]

If I have the same variable in the same population (qualitative, like smoking habit: 1:yes 2:no 3: ex) and I want to see how it evolves on time (baseline-12 months-24 months) can I use Chi-Square?

[@amirezat7854]

Thank god for your existence!!

[@RUS-dv5sm]

I was told that for a 2x2 table we must consider the continuity correction rather than the Pearson's Chi-square, is that true?

[@kr050]

If you are happy to believe that a binomial distribution is better approximated by a normal distribution. Why not do both an see if it matters for your conclusions?

[@MoayyadWM]

How can you make SPSS shows words such as smoker or non-smoker instead of numbers?? I thought it only accepts data as number

[@Gheezje]

+Moayyad Malas you click on the "value labels" icon, if you look in the video, its the fourth icon from the right (with the A and 1 with one purple arrow pointing up and the other one pointing to the left)

[@LoveIsNotits]

What are the general rules to consider in using chi square?

[@alyssasali377]

How do I find the correlation between a nominal variable and ordinal variable?

[@alphabetsouptoday]

wait, what if you have ordinal variables? What test should you use? I thought chi square and fisher's exact test were ok if you also had ordinal variables

[@hectorponce2012]

What is the difference between chi square test of proportions vs. chi square of independence?

[@m.hasnatakhtar187]

my test results are showing a negative Phi value. ik its a negative correlation but how can I relate it to my variables if someone knows the video link for an explanation. help me I'm stuck

[@ErayErtugrul09]

Perfect explanation!

[@kaiivarmuller8624]

Thank you very much for the example, but should`nt you read the value from the second row (Continuity Correction)?

[@anuragjena9007]

Thanks for the video But how do I deal with data in which I have examined around 4000 animals for a specific disease

[@pietromaximoff2363]

i have 2 variables.independent is smoking and gender,i also have to add frequency.is frequency gonna be measured as scale? because the other 2 are nominal

[@fffffmenicafffff]

i watched a few videos about this and you are the only one who said that chi square is used only for nominal variables.. what should we use for ordinal variables? or scale?

[@MrPetur4o]

Excellent video, thanks!

[@amalik1171]

We can use chi square for categorical variable

[@ksizz13]

How did you put in the values ?

[@ТетянаЮріївнаБабич]

Thank you for your helpful information!

[@tg5175]

can i perform chi sqare test betwee a nominal and an ordinal (Likert) variable?

[@aymansgarage]

What do you do if it has been violated?

[@Katiuska_F09]

If one of my variables is ordinal and the other is nominal, can I still use crosstabs and chi-square test?

[@jellehuvers4143]

Do you know already, because i also wanna know

[@yiingyiingSF9]

Hi what should I do if 25% of my data is less than the expected count?

[@noemiferrando5058]

thank you very good explained and good timing too

[@walther89]

how do you explane that result then ? at cramer v, by sentences

[@jeysy7902]

How its significant and the pval>0.05?? Plz sm1 answer me

[@levinabenita3487]

Yes i was wondering the same. For the result to be significant the p value has to be

[@Quadruple_curious]

What do you mean? the p value she got was .035, which is less than .05

[@nikeforo2612]

Thank you for the video, it's very easy to follow. I have tried to use the same 'style' in my book 'From Data to Insights: A Beginner's Guide to Cross-Tabulation Analysis' which will be published in July (2024) with Routledge. Perhaps it might interest you. Regards.

[@manalahmadsamamahmas2213]

Does this tea show if correlation is negative or positive... thank you

[@manalahmadsamamahmas2213]

Test I mean

[@careair5516]

Why is a chi-square use for statistical independence test? Convolution of two uniformly distributed r.v. (assuming maximum entropy) leads to a chi-square distribution? Please advise. Thank you for your assistance. 😃😃😃