banjalukaforum.com

Dobrodošli na banjalukaforum.com
Danas je 28 Mar 2024, 14:19

Sva vremena su u UTC [ DST ]




Započni novu temu Odgovori na temu  [ 8 Posta ] 
Autoru Poruka
 Tema posta: Statistika
PostPoslato: 05 Avg 2013, 21:26 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 21 Jan 2006, 22:41
Postovi: 1995
Lokacija: Izmedju zemlje i neba
Prijavio sam online ucenje statistike i sada je stigao zavrsi ispit i od njega mi zavisi ocjena. Nisam bas savladao celo gradivo pa bi bio zahvalan za bilo kakvu pomoc. Bez obzira na polozenu statistiku na fakultetu i zelje da iskoristim ovaj online kurs za podsjecanje ovo je za mene kao nuklearna fizika, vecinu stvari ne razumem bez obzira sto sam odslusao predavanja. Postavicu ovde pitanja a vi pomozite u odgovorima. Hvala unapred. Taman i dobra prilika za vas da testirate vasa znanja iz statistike.
Znaci u pitanju je multiple choise.





1. If scores on an exam are normally distributed with a mean of 80% and a standard deviation of 10%, what would a score of 90% be in Z-scores (e.g., how many standard deviations is this score above or below the mean)?
Z=-1.5
Z=-1.0
Z=0
Z=1.0
Z=1.5


2. Which of the following is a measure of variability?
Median.
Interquartile range.
Correlation coefficient.
Mean.
Risk ratio.



3. A particular diagnostic test for disease X has a sensitivity of 90% and a specificity of 90%. What is the negative predictive value of the test?
90%
10%
1%
18%
It cannot be determined from the information given.



4. In a randomized trial of two drugs to treat depression (drug A and drug B), depressive symptoms decreased significantly in patients on drug A (p<.001) but not in patients on drug B (p>.05). It follows that:
Drug A is superior to drug B.
Drug A and drug B are equally effective.
Drug B is superior to drug A.
None of the above.



5. In a psychology experiment in which 100 volunteers were asked to read a paragraph about an engineer, 65 assumed that the engineer was male despite the fact that the paragraph did not specify gender (and avoided gendered pronouns such as “he” or “she”). If the null hypothesis here is that there is no gender bias, what is the two-sided p-value associated with this result? Use a normal approximation to solve this.
P=.0001
p=.003
p=.10
p=.05
p=.99


6. Researchers reported that in a sample of U.S. women aged 50 to 54 who underwent mammography, 14.4% were recalled for further evaluation; however, in a similar sample of women undergoing mammography in the United Kingdom, only 7.6% were recalled for further evaluation.

What are the observed risk and odds ratios for the association between living in the United States (“exposed”) vs. the United Kingdom (“unexposed”) and being recalled for further evaluation following mammography (outcome)?
RR=1.89; OR=2.05
RR=2.05; OR=1.89
RR=0.40; OR=0.30
RR=0.30; OR=0.40
RR=2.05; OR=3.0



Use the following figure to answer questions 12-14: Figure 3b (British J Derm 2008) comes from a randomized trial of DHA versus a placebo pill for treating eczema. The figure shows boxplots of the eczema severity score, SCORAD, at baseline and week 8 for each group. SCORAD score is not normally distributed and the sample size is small.
Slika

The image shows 4 box plots. Box Plot A (DHA, Week 0) Median 32, lower quartile 29 and upper quartile 44 (minimum 18 and maximum 49); Box Plot B (DHA, Week 8) Median 28, lower quartile 23 and upper quartile 34 (minimum 18 and maximum 52); Box Plot C (Control, Week 0) Median 36, lower quartile 30 and upper quartile 45 (minimum 18 and maximum 65); Box Plot D (Control, Week 8) Median 34, lower quartile 25 and upper quartile 38 (minimum 13 and maximum 55)


12. What test should be used to determine whether DHA is a better treatment than control for treating eczema (SCORAD is not normally distributed)?
A t-test that compares A and B.
A Wilcoxon sign-rank test that compares B and D.
Wilcoxon sign-rank tests comparing A with B and C with D.
A Wilcoxon sum-rank test that compares the difference between A and B with the difference between C and D.
A chi-square test that compares B and C.


13. Use the graph above to roughly estimate the standard deviation of SCORAD score in the DHA group at baseline and the control group at baseline.
DHA group = 50, control group=100
DHA group=100, control group=50
DHA group = 6, control group=8
DHA group = 8, control group=6



14. What does the box in a box-plot represent?
Everyone who is not an outlier goes in the box.
The middle 50% of the data.
The mean +/- 1 standard deviation
The mean +/- 2 standard deviation
The part of the data that are normally distributed


15. You are performing a cohort study to look at risk factors for breast cancer in postmenopausal women. If the probability of developing breast cancer among smokers is .05 for the study duration, then if you sample 100 smokers, how many do you expect to develop the disease? Give the expected value +/- 1 standard deviation.
5±60
10±6
5±2
10±5
5±6


16. A study examining the relationship between fetal X-ray exposure and a particular type of childhood blood cancer found the following odds ratio (and 95% confidence interval) for the association: 2.44 (0.95 to 6.33). This result would likely be considered:
clinically significant, but not statistically insignificant
neither clinically nor statistically significant
both clinically and statistically significant
clinically insignificant, but statistically significant



17. Standard error:
Is always the standard deviation divided by the square root of n.
Is a measure of the variability of a sample statistic.
Increases with bigger sample sizes.
All of the above.


18. Which of the following would increase the width of a confidence interval?
Changing from a 99% to 95% confidence level.
Increasing the variability of the outcome.
Increasing the sample size.
Removing an outlier from the data.



Use Figure 3 to answer questions 19-20:
Slika
The following figure is from a study of exercise endurance (duration on an exercise test) in a group of 13 children with cystic fibrosis before and after an intervention program (a 4-week summer camp). For example, subject 1 improved by 3 minutes. Figure 3. Exercise endurance (duration) before and after camp. The image show a bar chart. After the summer camp, Subject 1 improved from 4.5 to 7.5 mins; Subject 2 improved from 7.0 to 7.5 mins; Subject 3 remained at 10.5 mins before and after the camp; Subject 4 improved from 7.5 to 8.0 mins; Subject 5 improved from 6.5 to 8.0 mins; Subject 6 remained at 6.5 mins before and after the camp; Subject 7 improved from 6.0 to 7.5 mins; Subject 8 improved from 8.5 to 9.0 mins; Subject 9 improved from 7.0 to 7.5 mins; Subject 10 improved from 5.5 to 6.5 mins; Subject 11 improved from 8.0 to 8.5 mins; Subject 12 remained at 7.5 mins before and after the camp; Subject 13 improved from 3.5 to 4.0 mins.



19. What is an appropriate statistical test for analyzing these data assuming that the change in exercise duration is normally distributed?
two-sample ttest
paired ttest
ANOVA McNemar’s exact test
Fisher’s exact test


20. Suppose that the researchers had included a control group of 13 cystic fibrosis patients who did not attend the interventional camp. What would be an appropriate statistical test for comparing the improvements in the two groups (assuming that the change in exercise duration is normally distributed)?
two-sample ttest
paired ttest
chi-square test
logistic regression
Cox regression


21. A study was conducted to examine the peer review process. The investigators hypothesized that reviewers suggested by authors would give more favorable reviews than reviewers picked by journal editors. They obtained data on 40 manuscripts that had been reviewed by 1 author-suggested and 1 editor-suggested reviewer. They obtained the following results:
Author-suggested reviewer
Editor-suggested reviewer Favorable (accept/revise) Unfavorable (reject)
Favorable (accept/revise) 10 1
Unfavorable (reject) 9 20

Calculate the exact two-sided p-value associated with this outcome (calculate the exact binomial probability).
.021
.043
less than .0001
.51
.01


22. In a survey of 1000 registered U.S. voters, 55% respond that they support health care reform. What is the 95% confidence interval for the true percentage of U.S. registered voters who support health care reform?
50%-60%
55%-60%
51%-59%
52%-68%
49%-47%



24. Assume that the probability of developing lung cancer in smokers is 15%; the probability of developing lung cancer in non-smokers is 1%; and the prevalence of smokers in the U.S. is 20%. If a person is diagnosed with lung cancer, what is the probability that he/she is a smoker?
38%
79%
89%
95%
50%




25. In a cross-sectional study of heart disease and gender in middle-aged men and women, 10% of men in the sample had prevalent heart disease compared with only 5% of women in the sample. After adjusting for age in multivariate logistic regression, the odds ratio for heart disease comparing males to females was 1.1 (95% confidence interval: 0.79—1.43). What conclusions can you draw?
Being male increases your risk of heart disease.
Age is a confounder of the relationship between gender and heart disease.
The men in the study are younger than the women in the study.
There is a statistically significant association between gender and heart disease.
The study had insufficient power to detect an effect.


26. Researchers conducted a case-control study to identify risk factors for kidney cancer. They asked 50 cases and 50 controls about 100 different exposures and personal characteristics, and calculated the odds ratio for kidney cancer for each risk factor. They found statistically significant odds ratios with 2 factors: coffee intake (p=.03) and cell phone usage (p=.04). The authors should conclude that:
Coffee and cell-phone usage increase the risk of kidney cancer.
The risks of coffee and cell-phone usage have been exaggerated due to the use of odds ratios.
These associations are likely chance findings.
The study had insufficient statistical power.
Coffee and cell-phone are statistically significant but not clinically significant risk factors for kidney cancer.


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 13 Jan 2014, 00:53 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 21 Jan 2006, 22:41
Postovi: 1995
Lokacija: Izmedju zemlje i neba
Ima li iko od forumasa da se razume u statistiku i da zna kako se interpretiraju rezultati Bland altmanovog plota i regresione analize. Uporedjivao sam dve razlicite metode i sada kada sam dobio rezultate nisam 100% siguran da znam sve procitati kako treba. Bojim se da ne izvucem krive zakljucke


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 13 Jan 2014, 03:06 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 29 Jun 2010, 16:23
Postovi: 3551
Lokacija: Kad kreneš od Boske ka mljekari
Kontaktiraj Archangel-a. Mislim da ti on može pomoći oko toga.

_________________
Ako si najbolji u sobi u pogresnoj si sobi


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 09 Feb 2016, 23:48 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 21 Jan 2006, 22:41
Postovi: 1995
Lokacija: Izmedju zemlje i neba
Imam opet problem sa statistikom pa ako znate nekoga da se razume da mi pomognete. Radio sam neka istrazivanja i mjerenja nekoliko parametara pa sam hteo da istrazim korelaciju izmedju dva. Rad sam poslao na adresu jednog casopisa koji mi je odgovorio sledece:

3. There are no metrics for the reliability of measures. The Journal's preference is the one advocated by the International Society for the Advancement of Kinanthropometry: technical error of measurement.

6. While you state confidence intervals, you do not fully engage with magnitude-based inferences.

7. Relationships are evaluated on the basis of correlation coefficients. It is well established that in bivariate analyses, these coefficients can be influenced more by the range in one or both variables than inherent relationship between the two.

Oni sad traze da ispravim ovo u radu pa da prihvate rad za publikaciju. Iskreno sa statistickog aspekta nemam pojma sta zele pa bih bio zahvalan na pomoci


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 10 Feb 2016, 08:30 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 01 Mar 2012, 10:48
Postovi: 1852
Lokacija: Transilvanija
3. Za to sto si mjerio postoje ustanovljeni standardi kvaliteta vezani za metode mjerenja (odnosno, ono sto forsira International Society for the Advancement of Kinanthropometry). Potrebno je da kazes da li si ih ispostovao ili obrazlozis zasto nisi, ukoliko nisi. A ako jesi postovao standarde, traze konkretno mjerenje kvaliteta izvrsenih mjerenja. Ne znam da li ovo ima veze sa statistikom (jedino ako je u pitanju neka statisticka metoda za mjerenje kvaliteta mjerenja).

6. Ne kontam ovo. Izracunao si intervale povjerenja za nesto, ali za to isto ti nedostaju magnitude-based inferences, sta god to bilo. Drugo tumacenje je, da si izracunao intervale povjerenja za nesto, ali nisi bas potpuno smisleno to uradio (to je varijanta da izracunavanje intervala povjerenja zapravo vid magnitude-based inferences).

7. Koeficijenti korelacije nisu bas adekvatan metod za shvatanje odnosa izmedju varijabli koje si posmatrao (izracunao koralaciju za njih). Mislim da ti sugerisu da razmotris upotrebu adekvatnog regresionog modela, pored korelacije, s ciljem objasnjavanja odnosa izmedju varijabli.


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 10 Feb 2016, 11:10 
OffLine
Samo jedan iznad mene
Samo jedan iznad mene
Korisnikov avatar

Pridružio se: 03 Jan 2007, 16:38
Postovi: 23477
Drakula da li je tvoj nik dr Akula? Pošto vidim da poznaješ doktorski nivo statistike :)

_________________
Rad je od čovjeka napravio roba, neradu smo odani sve do groba.


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 16 Mar 2016, 07:39 
OffLine
Veteran
Veteran
Korisnikov avatar

Pridružio se: 21 Jan 2006, 22:41
Postovi: 1995
Lokacija: Izmedju zemlje i neba
Ni na ovo pitanje vezano za statistiku nisam uspeo naci odgovor. Ni jedan statisticar u BL mi nije mogao pomoci. Sta reći, kakvu poruku poslati?


Vrh
 Profil  
 
 Tema posta: Re: Statistika
PostPoslato: 13 Jul 2019, 17:08 
OffLine
Pripravnik
Pripravnik

Pridružio se: 29 Maj 2003, 23:48
Postovi: 158
Lokacija: tunjice
Da provjerim jesam li još živ ovdje.
"3. There are no metrics for the reliability of measures. The Journal's preference is the one advocated by the International Society for the Advancement of Kinanthropometry: technical error of measurement."
Otprilike:3. Ne postoje podaci za pouzdanost mjerenja. Časopis daje prednost za ono što zagovara Međunarodno društv za unapređivanje kinantropometrije: tehnička pogreška mjerenja.

Naprimjer važeš iste jabuke na dvije različite vage,polužni kantar i neku elektronsku.
Kantar kaže 2 kg +/-10 gr,sa pouzdanošću 90%,ili 2 kg+/- 20 gr sa pouzdanošću 99%
El.vaga kaže 2 kg +/- 1 gr.sa pouzdanošću 66%,ili 2kg +/- 3 gr. sa pouzdanošću 90%.

Taćnost za datu pouzdanost je karakteristika mjerne opreme i neophodna je za dalju statističku obradu.

_________________
cuvajte se lopova!


Vrh
 Profil  
 
Prikaži postove u poslednjih:  Poređaj po  
Započni novu temu Odgovori na temu  [ 8 Posta ] 

Sva vremena su u UTC [ DST ]


Ko je OnLine

Korisnici koji su trenutno na forumu: Nema registrovanih korisnika i 3 gostiju


Ne možete postavljati nove teme u ovom forumu
Ne možete odgovarati na teme u ovom forumu
Ne možete monjati vaše postove u ovom forumu
Ne možete brisati vaše postove u ovom forumu
Ne možete slati prikačene fajlove u ovom forumu

Pronađi:
Idi na:  
Powered by phpBB® Forum Software © phpBB Group
Hosting BitLab
Prevod - www.CyberCom.rs