TEST OF SIGNIFICANCE.pptx

dentistry ,biostatistics, periodontics

1. TEST OF SIGNIFICANCE Dr Joice P Jiji MDS first year GUIDED BY Dr Sandeep J N

2. contents Introduction Common terms in statistics Biostatistics Hypothesis –null and alternate Test of significance Parametric and non parametric tests Z- test Student t- test ANOVA Chi Square test

3. Mc Nemar’s test The Wilcoxon signed-rank test Mann –Whitney U test The Kruskal Wallis test The Friedman tests   Fisher's exact test   Post hoc rank test Conclusion references

4. Introduction Statistics  is the discipline that concerns the collection, organization, analysis, interpretation, and presentation of data Statistics is an important and integral part of research methodology. It is a pervasive force on which the entire spectrum of clinical decision making is dependent

5. Test of significance are one of the central concept in statistics These are mathematical method by which the probability of an observed difference occurring by chance is found.

6. Common terms in statistics

7. Variable : A characteristic that takes on different values in different, places or things. 2 types dependent and independent variable A dependent variable is the  variable  being tested in a scientific experiment. Examples of independent variables are • Age, sex, race

8. Population : It is an entire group of people or study elements— persons, things or measurements for which we have an interest at a particular time. Populations are determined by our sphere of interest. It may be infinite or finite.

9. Sampling unit : Each member of a population. Sample : It may be defined as a part of a population. It is a group of sampling units that form part of a population, generally selected so as to be representative of the population whose variables are under study. There are many kinds of sample methods in Biostatistics.

10. Mean This measure implies arithmetic average or arithmetic mean which is obtained by summing up all the observations and dividing the total by the number of observations. Median When all the observations of a variable are arranged in either ascending or descending order, the middle observation is known as median Mode This is the most frequently occurring observation in a se

11. 6mm,7 mm,4mm, 6mm,5 mm,6 mm,8 mm,4 mm Mean = 6+ 7+4+6+5+6+8 +4 = 46/8 = 5.75 mm 8 Median = 4mm,4mm,5mm, 6mm,6mm ,6mm,7mm,8mm Median 6+6 = 12/2= 6 mm 2 Mode= 6mm 4mm,4mm,5mm, 6mm,6mm ,6mm ,7mm,8mm

12. Normal distribution and Normal curve  Gaussian distribution, First observed by Abraham de Moivre in 1733 Probability distribution that is symmetric about the mean. A theoretical , continuous, symmetrical ,unimodal distribution of infinite range Most of the biological variables follow normal distribution

13. The characteristics of a normal curve are: 1. It is bell-shaped 2. It is symmetrical. 3. Mean, mode and median coincide. =0 4. It has two inflections Total area is =1 Standard deviation=1

14. STATISTICAL INFERENCE Inference means drawing of conclusion from data It is the process of drawing up conclusions from quantitative or qualitative information using the methods of statistics to describe and arrange the data and to test suitable hypothesis Statistical inference are based on probabilities and as such cannot be expressed with full certainty

15. BIOSTATISTICS Biostatistics is the term used when tools of statistics are applied to the data that is derived from biologic science such as medicine Statistical analysis is the back bone of research

16. TEST OF SIGNIFICANCE Whenever two sets of observation are compared, it become essential to find out whether difference observed between the two group is because of sampling variation or any other factor

17. STAGES IN PERFORMING A TEST OF SIGNIFICANCE 1 Create a null hypothesis 2 Create an alternative hypothesis 3 Determine the significance level 4 Decide on the test we will use 5 Perform a power analysis to find out your sample size

18. 6 calculate the standard deviation 7 Use the standard deviation 8 Determine the t- score(F,H,P,U) 9 Find the degree of freedom 10.Use a t- table for determining p value

19. In  Statistics, a hypothesis is defined as a formal statement, which gives the explanation about the relationship between the two or more variables of the specified population. It helps the researcher to translate the given problem to a clear explanation for the outcome of the study. What is a hypothesis ??? 20

20. CHARECTERISTICS OF HYPOTHESIS Hypothesis should be clear and precise. Hypothesis should be capable of being tested. It should state relationship between variables.

21. 4 It must be specific. It should be stated as simple as possible. It should be amenable to testing within a reasonable time. It should be consistent with known facts.

22. classification Based on their formulation Null hypothesis and alternate hypothesis

23. Null hypothesis It states that there is no real difference between the means (or proportions ) of the groups being compared (or that there is no real association between two continuous variables). It is denoted by Ho. Example- “ There is no difference in the clinical attachment level with treatment A or treatment B”.

24. Step 2 Alternate hypothesis If null hypothesis is rejected, we need another hypothesis…..an alternate hypothesis. It states that there must be a true difference between the groups being compared. It is denoted by HA or Ha or H1. Example- “there is a difference in the clinical attachment gain with treatment A and treatment B”.

25. In Hypothesis testing we proceed on the basis of Null Hypothesis. We always keep Alternative Hypothesis in mind. It seems strange to begin the process by asserting that something is not true, but it is far easier to disprove an assertion than to prove that something is true. The Null Hypothesis and the Alternative Hypothesis are chosen before the sample is drawn.

26. Step 3 Determining LEVEL OF SIGNIFICANCE Before the study is started, we have to establish a criterion called level of significance or alpha level which is the highest risk of making a false positive error of rejecting the null hypothesis that the investigator is willing to accept. So the confidence with which the null hypothesis is rejected or accepted is called as Level of significance A common alpha is 0.05 or 5%

27. The higher the significance level used for testing a hypothesis, the higher the probability of rejecting null hypothesis, when it is true. If P value is less than alpha ,we will reject null hypothesis If P value is more than or equal to alpha we will not reject null hypothesis

28. STEP 4 Decide which test we have to use? Parametric tests Their model specifies certain condition about the parameters of the population from the research sample is drawn Used for quantitative data Non- parametric tests Their model does not specify condition about the parameters of which the research sample is drawn Used for qualitative data

29. Quantitative (or Continuous) Data The quantitative data have a magnitude. The characteristics is measured either on an interval or on a ratio scale. It got a numerical value Age Height, weight RAL

30. Qualitative (or Discrete) Data In such data there is no notion of magnitude or size of the characteristic or attribute as the same cannot be measured. Young and old Gender Social class Redness of gingiva Efficacy of drug

31. Parametric tests Non parametric test Paired t-test Wilcoxon signed rank test Unpaired t-test Z test Mann-Whitney U-test Chi square test One way ANOVA Kruskal Wallis test Fischer exact probability test Repeated ANOVA Friedman test Mc Nemar’s test

32. 1 large sample tests When the sample size is greater than 30 Generally ,2 types of data may be encountered while testing hypothesis for large samples When data is qualitative -------- test for proportion( chi square /x 2 test) When data is quantitative -------- test for means(z-test )

33. 2. small sample tests W hen the sample size is smaller than 30 Sample does not follow the normal distribution ,hence it is based on the assumption that the population from which the sample is drawn follows the normal distribution Student T test unpaired and paired ANOVA Chi-square test (pronounced as kye)

34. Choice of an appropriate statistical significance test to be used Association between two variables---- chi-square test Correlation between two variables ----- pearson’s or spearman’s test One group on two occasions-------paired test One group on 3 or more occasions------------------ ANOVA Two separate groups------------unpaired t test, Mann-Whitney u test 3 or more separate groups--------ANOVA

35. step 6- Standard deviation The standard deviation is the most important and widely used measure of studying dispersion. It is also known as root mean square deviation Small standard deviation means a higher degree of uniformity of observation Standard deviation expressed with sigma s

36. Formula to caculate standard deviation s 6mm,7 mm,4mm, 6mm,5 mm,6 mm,8 mm,4 mm Mean = 6+ 7+4+6+5+6+8 +4 = 46/8 = 5.75 mm 8 SD s = 1.38873

37. Student’s t-test Designed by W.S Gosette , whose pen name was Student t is ratio of observed difference between two means of small samples to the standard error of difference in the same Sample here should be less than 30

38. t= difference between two means S.E of difference between two means T test Paired t test Unpaired t test (independent or unmatched or pooled t test)

39. Paired test- e.g clinical attachment loss(CAL) before and after scaling Unpaired t-test e.g. effect or scaling in males and females , The mean PI, GI, PD & CAL between 02 groups at different time intervals.

40. UNPAIRED T TEST-Steps 1 Hypothesis 2 find the observation difference between means of two samples (x 1 -x 2 ) 3 calculate the standard error (SE) of difference between the two means 4 . Calculate t value t= (x 1 -x 2 ) SE SE= s root of 1/n 1 + 1/n 2 s - standard deviation

41. PAIRED T TEST-Steps 1 null hypothesis 2 find the observed difference in each set paired observations before and after of the same sample(x 1 -x 2 =x) 3 calculate mean of the differences 4 workout the standard error of the mean 5. Calculate the t value

42. Z test (standard normal test) z test is done when the population is more than 30 for quantitative data Used to compare : Two sample means Sample mean with population mean Two sample proportions Sample proportion with population proportion

43. X- sample average m - mean s - standard deviation

44. AN O VA Analysis of variance(F test) Developed by Professor R A Fisher Analysis of variance is useful to assess the significance of difference of differences between sample means which are more than two in number

45. If the independant variable quantitative and categorical (i.e. nominal, dichomatous , ordinal) the correct multivariable technique is ANOVA

46. One way ANOVA If the design include one independent variable that technique is called one way ANOVA, regardless of how many different groups are compared. Another term for one-way ANOVA is F- test. F-test is a kind of super t-test that allows the investigator to allow more than two means simultaneously.

47. The ratio of the between group variance to the within groups variance is called F(in honor of Fischer). F= s 2 1 based on variation between the group s 2 2 based on variation between the group

48. Two way ANOVA More than one independent variable present eg treatment plan A and B , age, sex The goal of ANOVA is to explain as much variation in the continuous variable as possible, by using one or more categorical variables to predict the variation……….. In this the impact of two different factors on the variations in a specific variable is tested. Two way ANOVA is also called N way ANOVA

49. The chi square test for quantitative data Chi square test is used for comparing a sample variance to population variance x 2 = s 2 s (n-1) s 2 p

50. Non parametric test Chi square test for qualitative data ,developed by karl pearson Used to test the association between two events (to test a given hypothesis) e.g. cause and effect like tobacco use and cancer Χ 2 = ∑ (Oi – Ei) 2 /Ei Oi observed frequencies Ei excepted frequencies Frequency means the no of times the value occurrs in the data

51. P aired samples--The Wilcoxon signed-rank test Also known as matched pair test It is  a non-parametric statistical hypothesis test used either to test the location of a population based on a sample of data, or to compare the locations of two populations using two matched samples . Like more aggressive and less aggressive

52. 3.unpaired samples -Mann –Whitney U test A non parametric test used to compare the medians of two independent sample.it is the non parametric equivalent of the t test e.g , GCF GF levels between 2 groups at different time interval

53. U=n 1 n 2 + n 1 (n 1 +n 2 ) - R 1 2 U=Mann-Whitney U test n 1  = sample size one n 2= Sample size two R i  = Rank of the sample size

54. 4. Fisher's exact test  Fisher's exact test  used in the analysis of  contingency tables . Although in practice it is employed when  sample  sizes are small, it is valid for all sample sizes. Inventor -   Ronald Fisher , and is one of a class of  exact tests

55. Success and failure in group 1 and 2

56. 5.McNemar’s test A variant of chi squared test ,used when the data is paired it can be used to analyze retrospective case-control studies, where each case is matched to a particular control. Or it can be used to analyze experimental studies, where the two treatments are given to matched subjects.

57. 6.The Kruskal Wallis test The Kruskal Wallis test is the non parametric alternative to the One Way  ANOVA .  The H test is used when the assumptions for ANOVA aren’t met (like the  assumption of normality ). It is sometimes called the  one-way ANOVA on ranks , as the ranks of the data values are used in the test rather than the actual data points. The test determines whether the  medians  of two or more groups are different.

58. 7. The Friedman test developed by  Milton Friedman The Friedman test is used for one-way repeated measures analysis of variance by ranks. In its use of ranks it is similar to the  Kruskal–Wallis one-way analysis of variance  by ranks. The Friedman test is widely supported by many  statistical software packages .

59. Post-hoc test (multiple comparisons) “AFTER THIS” For comparison of three or more group means we apply the analysis of variance (ANOVA) method to decide if all means are equal or there is at least one means are equal or there is at least one mean which is different from others. If we get significant result we can conclude that there is difference in group means To know what specific pairs of group means show differences- Post-hoc test (multiple comparisons) procedures.

60. The set of comparison is referred as a family of test Bonferroni correction- safe option Turkey’s HSD procedures- assumption met Scheffe's procedures Newman – keuls procedures Dunnette’s procedures

61. Step 9 calculating Degree of freedom No of independent members in the sample The degrees of freedom formula is straightforward. Calculating the degrees of freedom is often the sample size minus the number of parameters you’re estimating:

62. step10.Use a t- table for determining p value

63. The level of statistical significance is often expressed as a p -value between 0 and 1. The smaller the p-value, the stronger the evidence that you should reject the null hypothesis. A p -value less than 0.05 (typically 0.05) is statistically significant. If P value is less than alpha ,we will reject null hypothesis If P value is more than or equal to alpha we will not reject null hypothesis

64. Computer in biostatical analysis The MINITAB , SPSS , SAS and STATA are the some well known statistical software packages for personal computer which are used for the tabulation and statistical analysis of data SPSS( Statistical Package for the Social Sciences) is commonly used

65. Limitations of test of significance 1.Test are only useful aids for decision making not decision making itself 2.Do not explain why does the difference exist 3. Result are based on probabilities and such cannot expressed in full certainty 4.Inferences based on them cannot be said to be entirely correct evidence connecting truth of the hypothesis

66. Conclusion Tests of significance play an important role in conveying the results of any research and thus the choice of an appropriate statistical test is very important aa it decides the fate of out come of the study The tests are only useful aids for decision-making. Hence “proper interpretation of statistical evidence is important to intelligent decisions.” If the data deviate strongly from the assumptions of a parametric procedure, using the parametric procedure could lead to incorrect conclusions.

67. References Soben Peter Community Dentistry 6th Edition. Text book methods in bio statistics 7 th edt . B K Majajan Jain S, Gupta A, Jain D. Common statistical tests in dental research. Journal of Advanced Medical and Dental Sciences Research. 2015 Jul 1;3(3):38. Joseph john Preventive and community dentistry 2 nd edt

68. My sincere thanks to DR ARUNA D R DR VINAYAK S GOWDA DR AVINASH J L DR RAJIV N P DR REKHA JAGADEESH DR SANDEEP J N DR SOWMYA PRAVEEN ALL MY POST GRADUATE COLLEGUES

69. Previous seminar questions COMPONENTS OF LA Local anaesthetic drug ----lignocaine hydrochloride Vasopressor/vasoconstrictor drug---------Adrenalin Preservatives-methylparaben, thymol, chlorbutol Sodium chloride/Ringer’s solution Distilled water General preservative Alternative for methylparaben---sodium bisulfite , metabisulfite

70. Short acting corticosteroid

71. Anaphylaxis- treatment The first thing to do is to stop injection of allergen. Call for help, check patient's vitals Use 0.5ml ml of 0.1% of epinephrine intravenously. If the patient pressure would not stabilize after 15 minutes, repeat this procedure. max 3 times corticosteroids are very useful for such cases. IV OR IM (prednisone, dexamethasone and hydrocortisone) If a patient has signs of asphyxiation you should make an injection of aminophylline 2.4% – 10-20 ml intravenously Shift the patient to near hospital as early as possible

72. Signs and symptoms of anaphylactic shock feeling lightheaded or faint. breathing difficulties – such as fast, shallow breathing. wheezing. a fast heartbeat. clammy skin. confusion and anxiety. collapsing or losing consciousness. hypotension Skin reactions, including hives and itching and flushed or pale skin.

73. Blood loss during flap surgery Ariaudo 1970 observed that a full mouth flap surgery under general anesthesia blood loss is around 300 ml. According to D A Baab 59.47+or-38.2ml Berdon -gingivectomies involving 5 to 14 teeth -5 ml to 149 ml According to Mclvor and Wengraf -10 fold increase in blood loss per tooth during periodontal flap than gingivectomies

74. Other names of lignocaine Lidocaine Brand name xylocaine