Kenapa n=30 utk pilot/experimental test?

Selama ni tau 30 jek, sebab musabab x pasti..ini penjelasannya, terima kasih Dr OT.

mcm mana nak interpret t test spss output

As salam...ni pon topik basi ni..tp nak masuk jugak kat sini..sbb bila lama tinggal..mampuih lupa semua..setiap kali sambung blaja, setiap kali repeat balik mende ni..lupa sbb x amalkan..yelah...kita buka guna spss nak beli barang kat tesco kan..hahahahahah...

Ckp pasal t test kan..kompom2 la statistik kan..selalunya org xde masalah nak dptkan result guna spss..pastu..nak interpret...menggagau x reti..spss tu bagi result dlm bentuk nombor..kalau x reti baca maksud nombor tu..xde maknanya...

basicnya t test ada 3 jenis..semua org slalunya dah familiar kan..simple t test..independent t test dan sekor lagi..dependent  test a.k.a paired t test...

sblm syok2 ber t test..pastikan data korg2 ni semuanya normally distributed eh..bell atau gaussian shape..bukan skewed ke kiri atau kanan..graph bentuk m ke v ke..bukan ek..itu kene dipastikan..how ler?? kau buat la NORMALITY TEST guna spss tu haa..... tp kalau tgk dari bacaan varians..pon dah boleh agak juga nanti data kita jenis mcm mana..normal atau not normally distributed...mende varians tu??

Varians tu ialah satu bacaan yang menunjukkan mcm mana data tu disperse/bertabur/berselerak daripada min..eh...sebelum kau kompius..min tu purata..benda yg sama

makanya, kalau nilai varians kau besar..maknnaya lagi jauhla tersebar data2 tu daripada min..so..xde maknalah min itu..maksudnya lagi..sampel kau tu tidak homogenus..bila x homogenus maknanya not normally distributed..so xleh la pakai t test..sbb test ni parametrik, apply utk data normal saje..data tidak normal..silalah pi buat non parametrik test..ngerti??

t test ni guna utk tengok mean difference (perbezaan min)..jadi basicnya ko kene buat descriptive stats dulu..maknanya kene tau berapa mean, mod, varians semua2 tu..yg basic2..sbb ada kalau melalui descriptive stats..sah2 nampak ada mean different yg sagt besar dan jelas..x payah la gedik nak continue dgn t test..t test ni inferential stats..

cth output spss

kau tekan2 spss..keluar table ni..acaner nak baca??
1. ko tgk figure pada sig colume..tu levene test punya reading..sbb nilai levene test tu 0.582 (p>0.05), lebih besar dr significant 0.05..maka varians ada sama...maknanya kita akan amil bacaan equal variance (row atas) utk tgk t test value pulak..
2. kat ni t test value dia 0.14 (p>0.05), kita accept null hypotesis dengan rumusan bahawa tiada perbezaan min antara.....ko sambung ayat tu ikut variable kajian masing2 ye....

kalau mcm ni pulak???

sama aje la...
1. pi tgk kat Levene test bah sig tu..apsal?? sbb figure ni akan tentukan kita akan amik bacaan t test dr row atas atau bawah. Dalam kes ni, nilai sig Levene test ialah p=0.006 (p>0.05)..so..bacaan kita mesti dari row pertama.
2. bacaan t test..nampak 0.000..sah2 la p<0.005, so kita reject null hypothesis..dengan rumusan bahawa wujud perbezaan min antara.....ko sambung ayat tu ikut variable korg.....
perbezaan ini bermakna tau.. meaning...

cth: lelaki skor 50, perempuan skor 54
berapa beza?? 4 je kan...tp bila buat t test,,wujud perbezaan yg signifikan sbbP<0.05, so bermakna wujud perbezaan yg signifikan walau nilai kecil

senang kan!!
dah..kbye!

Tips for Defending Your Thesis

As Salam..

Share tips berguna utk kawan2 yg nak viva Isnin ni..seram dowh!

Kepada sape2 yg x tau viva tu apa..viva tu bukan kereta perodua viva yea, ini thesis defence..thesis oral examnination/presentation lebih kurang.. 

Ingat dah submit thesis tu ko dah selamat ker..belum lagi..next azab yg menunggu...kene viva..pemeriksa yg mark thesis ko akan ala2 menginterview ko..kerja2 yg ko dah buat, method yg diguna etc..

Aduhh..memikirkannya aje dah sakit perut! Mcm mana! Mcm mana!!

Rasa nak muntah ada..nak pengsan ada...nak membuang pon ada..semua ada secukup rasa...berdebar2 mcm nak kluar jantung..

Ya tuhan ...takutnya!!

Tips for Defending Your Thesis
By
Dr. Mani Maran Ratnam

Before the viva:

1. Read through your thesis once again and make sure that you 

    understand what you have written.
2. Read through the papers listed in your references so that you  

    understand the significance of your work in relation to other 

    researcher’s work.
3. Read through the relevant theories related to your thesis.
4. If you spot errors in your thesis, prepare an error list and the 

    replacement pages. Give it to the panel before the viva starts. 

    Apologize for the errors and don’t wait for the examiners to point 

    out the errors.
5. Write down the questions the examiners are likely to ask in each 

    chapter of your thesis. Prepare the answers to these questions. 

    Be prepared for more difficult questions like ‘What is the 

    significance of your findings?’, ‘How did you do your literature 

    search?’ etc.
6. Prepares slides for your presentation. Remember, you have to 

    present your two or three years’ work in 15 to 20 minutes! 

    Therefore, prepare well so that the examination panel can 

    understand what your problem is, how you solved it and what 

    you have achieved. Do not explain the theories in your 

    presentation. The examination panel do not want to listen to a 

    lecture! Make sure your slides consist of at least: Title page, 

    Objective of research, Past work (literature review),  

    Experimental and analysis procedure, Discussion of results, 

    Conclusion and Further work. Include a list of publications, if 

    you have published any part of the work. Use large font size (at 

    least 24) and do not put too much text on each slide. Choose 

    good contrast colors so that the texts stand out sharply from the 

    background. Do not use fancy background graphics or annoying 

    animations. Use plain background if possible. Use more 

    pictures and diagrams to make your presentation interesting. Do 

    not read from the slides! Prepare about 20 to 25 slides for a 

    20-minute presentation.
7. Burn your slides on a CD and bring it along, just in case your pen 

    drive or hard disk fails!
8. Remember to bring your copy of the thesis!! (latest version)

During the viva:

1. Come at least 30 minutes early and prepare your presentation. 

    Test the computer and projector so that everything works fine. 

     Dress properly. Remember, viva is a formal event.
2. Be relaxed when you start your presentation. Remember that you 

    know more about your work that the examiners or the panel 

    members. But, the examiners may know more about the 

    technical issues related to your field of research.
3. Understand the question fully before answering. Take your time 

    to answer the questions and restrict your answer to the question. 

    Be frank. If you are not sure of something, don’t defend it very 

    strongly. This will only encourage the examiners to dig further to 

    test your knowledge.
 

P/S: 

Ya Allah..bantulah mempermudahkan segala2nya..Semoga usaha selama ni akan membuahkan hasil..Saya mau konvo tahun nie..tak kira...tak kira... Utk kwn2 group MCN dan MCL yg akan menjalani viva this coming Monday (13th Feb 2012)..Good luck!

Kau rabunkanla sebentar mata2 depa..

As Salam...

Hii!!! weyyy..korg rindu aku x..hehehhehehe... 

Hhmm..Bokashi update blog yea hari tuh..xde keje la budak buntal debab kesayangan aku tuh...hikhik

 Beberapa hari lepas..mcm gambar tu la hidup aku..tonggang terbalik kelam kelibut huru hara sume cukup..bahana thesis kesayangan kononnya..hakhakhak...

x bernapas weh...sampai aku nak tdo pon rasa bersalah berdosa..sbb dok mengenangkan kerja2 terbengkalai..fuhhh...azab...azab..

 Tapi itula..bukan senang nak senang..x ada apa yg dtg bergolek..kene usaha..bukan ler nak kata aku giler tabah ataupon rajin tahap gaban..cuma mmg aku bersungguh2 la membuat benda nie...

Matlamatnya..no point buat thesis kalau xde value..kan..kan..

 Yang menambahkan sesak napas...bleh plak dlm kekalutan hidup tu..aku demam selsema batuk bagai..eh..eh..eh..dugaan betul..takpelah..menghapus dosa2 kecil..aku redha...dlm sakit2 tu gagah la gak buat..aku tau..kang bila dah siap..ini lah akan menjadikan kenangan manis...

 Ntah kenapa..hati x senang..makin dibaca makin byk spelling mistakes la..salah la..iskh..naik seriau..tension toksah citer..

Mmg aku kagum n tabik spring pd mereka2 yg dah settlekan phd..aku baru merangkak utk MSc ni pon dah rasa nak putus nyawa kadang2..mujur ada gak peneman setia, budak2 bulus dok teman buat keje sampai pagi..agak2 dah kelabu mata..aku gi la main sakat2 depa..tapi selalunya aku la kene sakat..bleh plak Bokashi debab tu sedap bantai tdo sebelah lappy..kalau si Lulu tu siap tersentil2 kluar lidah..tdo penuh perasaan gamaknya..geram je aku tgk..

Selain tu..aku juga sangat respect dgn kawan2 lain yg sama2 struggle dgn thesis..sbb ada di kalangan depa nih dah ada anak2 utk diurus..ada yg baru pas bersalin..

Ya Allah, Tuhan Maha Pemurah lg Maha Mengasihani..

Kau permudahkanlah urusan thesis kami2 nie..

Kau lembutkanlah hati para examiner supaya bagi markah kawkaw hebat kat kami..

Kau rabunkanla sebentar mata2 depa kalau terjumpa mana2 mistakes dan error..

Kau terbitkanlah rasa kesian dan kewajaran dlm pertimbangan semasa mereka sedang mark thesis kami..

Amin..Amin..Yarabbal Alamin...

P/S: Azab belum habis..baru separuh yg settle..hakhakhak..

BUSY DROWSY CRAZY WEEK

As Salam..

was BUSY yesterday, 

stil BUSY + DROWSY today, 

expected to be BUSY + CRAZY tomorrow..

Dear Allah, pls give strength to finish all these super BUSY DROWSY CRAZY stuff....

Life is hard..
InsyaAllah..will try my best to finish everything up..

Balik kg..Chapter 3..dan Owen..

As Salam..

Raya haji dah dekat..hhmm..busy la..tak sempat nak pikir apa..

Big Boss ajak balik kampung plak..haishh..aku dah ler tgh in the dark nih..rolling deep in my so called draft for chapter 3...kalau nak balik gak..terpaksa la ikut..akan ku angkut segala 'keje2 skolah' aku nih..dan jgn harap aku nak lepak2 silang kaki depan TV..bergurau senda & beramah mesra di pangkin...apatah lagi berdondang dendang dgn sudip dan periuk didapur..xdok masa....aku busy lah!! lain la kalau ada sape2 nak tolong tulis 'keje skolah' aku nih...ko ade??

 ohhh....jgn terkejut yea..kalau tgh wat keje mmg aku rupa ini..hehheheheh...takut tak?? takut takkkk???

Aktiviti peberet ialah tdo...dah bbrp hari tdo x lena..dlm tdo pon mimpi si lappy merahku memanggil2..SPSS dan data2 melambai2 sexy..apa kes?? kecik hati bantal busuk..sbb aku selalu menduakan dia ngan lappy...selimut plak dah lama x bertegur sapa dgn aku...hahahhaha..

 

Bila tgk anak2 bulus tdo..wah...wah...boleh meleleh air liur..giler jeles...dpt plak si Bokashi tdo sampai mengangkang2...hhmmm..tp xpela..bersusah2 dahulu..apapun mudah2an dipermudahkan urusan...

Dlm sibuk2 ni..hati2 yg tersayang juga harus ku jaga..kalau Big Boss kata balik kampung..menonong aku ikut je la..takkan nak derhaka plak...mak mertua aku pon suka kalau anak2 dia join dia balik kampung..sbb nampak gaya cucu2 x balik...whatever la..takmau dicop menantu derhaka juga..derhaka sana..derhaka sini...kang x cium bau syurga youuuu..takmo ler gitu...huhuhuhu...

Btw..lama x upload gambar2 anak2 bulus kan..ada gak yg tanya..

Bokashi mana?? 

Mak aji Betty sihat?? 

Rozy Tiger dah gemuk ke belum??

Terima kasih kpd yg bertanya..byk snap gambar depa..tp tak berkesempatan nak upload lg..Betty pon x leh nak update blog sbb mama dia pakai lappy most of the time..

Sebelum lupa..minta kawan2 doakan sikecil ni..dia menderita..mcm2 sakitnya..kenapa dia..siapa dia....sila jenguk wall FB Kucing Terbiar Anjing Jalanan (KTAJ)..

Nama budak ni OWEN..segala usaha..dari semudah dan semurah minyak gamat sampai ke ubat2 hebat dari vet dah dibuat..tp masih x mendatangkan hasil..selain virus..kidneynya juga dah rosak..mata sudah infection..Ya Allah..sengsara si comel ini..tp AWAS..jgn ada sape2 yg cadangkan PTS pada Owen..aku cili2..aku robek2 mulut ko nanti..walaupun dia derita...bukan hak kita utk memutuskan sama ada dia perlu hidup atau ditidurkan selamanya..jika dia masih survive ..pasti Allah ada sesuatu yg nak ditunjukkan..kita tak tau..kalau sudah sampai masa..Allah akan ambil dia..itu pasti...doakan Owen kawan2..pls send positive vibes to him..pd human parent si Owen..bertabahla kamu kawan..Insyaallah..

P/S: Gotta call HVC untuk boarding Mak aji Betty esok..sbb Big Boss punye idea nak balik kampung..sila sponsor boarding fees this time..kehkehkeh..

Parametric & Non-parametric Test

As Salam..

Me bz..stil struggling with my SPSS thingy..

Just a short note as a reminder:

1. always do TEST OF NORMALITY to check ur data distribution..normal or not normally distributed..this is important, it will help u to determine the type of test..either parametric test or non-parametric test..there's  a few item to look at..Shapiro-Wilk (for less than 100 sample) or Kolmogorov-Smirnov (for more than 100 sample), not to forget Q-Q Normal Plot and Boxplot.
2. For datas which are normally distributed...

• compare mean with reference/known value - simple student's T-test
• compare means between 2(only) independent group/sample ie: gender (male & female), diet (vege & non vege) - Independent T-test
• compare means between 2 dependent group/sample ie: before and 2 repetitive after value/readings/figure/results- Dependent T-test
• compare means between 3 or more group/sample ie: age range (10-15, 16-20, 21-25, 26-30..total 4 groups) - ANOVA (Analysis of variance) <--there are many types of ANOVA depending on the number of factors involved, one way or 2 way etc..gotta do more reading on this!! and Post-hoc test is needed if there is significant value..just to differentiate which groups/sample is actually giving significant result and which doesn't. Quite a number of post-hoc test..choose one that is suitable..regarding ur cases/sample.
• useful links for ANOVA:

http://www.psychstat.missouristate.edu/introbook/sbk27.htm

http://www.experiment-resources.com/anova-test.html

WHAT IF THE DATA IS NOT NORMALLY DISTRIBUTED??

That would be even easier!! We opt for non-parametric test..instead of comparing means..we compare median instead..so do your descriptive analysis before hand..

• for 2 different/independent group/sample - Mann Whitney Test
• for more than 2 group/sample - Kruskal Wallis Test
• for 2 dependent group/sample - Wilcoxon Test

How about correlation??

Normal distribution - Pearson Correlation

Not normally distributed - Spearman Correlation

P/S: Parametric tests are way better than the non-parametric..so if ur data is not normally distributed...just log(log/Ln) them...