[plural] discover —Research Process Tug CEU meo ead EI Reid einen cca Mecematosc Probie , 3 Ee even Tevet Perron ys TMs NRSGr Nee) Euchccltlas Tua elu) NAP enry feeer ten (ive toys Eee ea} V. Collect data t= Toe ol}Hypothesis Testing (a) Null hypothesis and alternative hypothesis: + In the context of statistical analysis, we often talk about null hypothesis and alternative hypothesis. * According to Fisher, a hypothesis which is tested for plausible rejection under the assumption that itis trueiscalledthe = Hypothesis’. * If we are to compare drug A with drug B about its efficacy and if we proceed on the assumption that both drugs are equally efficacious, then this assumption is termed as the null hypothesis. + Any other rival hypothesis iscalled 7Hypothesis Testing * The null hypothesis is generally symbolized as Hi and the alternative hypothesis as H.. * The null hypothesis and the alternative hypothesis are chosen before the sample is drawn . + Alternative hypothesis is usually the one which one wishes to prove and the null hypothesis is the one which one wishes to disprove. * Thus, a null hypothesis represents the hypothesis we are trying to reject, and alternative hypothesis represents all other possibilities.Hypothesis Testing (b) Type I error and type I error. + In the context of testing of hypotheses, there are basically two types of errors we can make. Bam Cmte nel (rows Pate cea yd Ce Deco Se Peete nartees ows Cniy rete Eom eis prog rieron yee oe Cott) et a If error. + Type I error means rejection of hypothesis which should have been accepted and Type II error means accepting the hypothesis which should have been rejected. * Type lL error is denoted by & (alpha) known as a error, also called the level of significance of test; and Type II error is denoted by B (beta) known as error.Hypothesis Testing The probability of committing a Type I error (chances of getting it wrong) is commonly referred to as p-value. * When a hypothesis test results in a p-value (probability of an a error) that is less than the significance level, the result of the hypothesis test is called statistically significant. + The conventional range for alpha is between 0.01 and 0.10. + Although numbers such as 0.10, 0.05 and 0.01 are values commonly used for alpha, there is no overriding mathematical theorem that says these are the only levels of significance that we EN monHypothesis Testing Why asl ean * ‘Type II error means accepting the hypothesis which should have been rejected. * The probability of a type II error is given by the Greek letter beta. * The probability of avoiding a type II error is called the power of the hypothesis test, and is denoted by the quantity 1 - £. + An attempt to decrease one type of error is accompanied in general by an increase in the other type of error. The only way to reduce both types of error is to increase the sample size, which may or may not be possible.Hypothesis TestingHypothesis TestingHypothesis Testing * The choice of significance level should be based on the consequences of Type I and Type II errors. + If the consequences of a type I error are serious , then a yery small significance level is appropriate. + Example 1: Two drugs are being compared for effectiveness in treating the same condition. Drug 1 is very affordable, but Drug 2 is extremely expensive. The null hypothesis is "both drugs are equally effective," and the alternate is "Drug 2 is more effective than Drug 1." In this situation, a Type | error would be deciding that Drug 2 is more effective, when in fact it is no better than Drug 1, but would cost the patient much more money. That would be undesirable, so a small significance level is warranted.Hypothesis Testing ava * Power is the probability of correctly rejecting a false null ih yeroldetoie *- Power=1—8 * Since Power is the probability of correctly rejecting a false null hypothesis, It is to our best interest to increase power. + There are several ways to increase power: — Increase the sample size. — Increase the alpha level. You will have more of a chance of rejecting the alternative at the 5% level of significance than a 1% test. — Consider an alternative that is farther away from the null hypothesis.X-p C—ph p}___A> A |_1_, Ssh a vn Vn G= Standard deviation n= Sample Size X= MeanHypothesis Testing Confidence interval: A is. a range of values within which the population parameter is expected to occur. The two confidence intervals that are used extensively are the 95% and the 99%. The upper and lower limit for a confidence interval is given by : ¥+z ae Tas Reema ent PATO AV CACO aT eer a Meta TOa) n= sample sizeCritical Values of z and Levels of ConfidenceThe Dean of the Business School wants to estimate the mean number of hours worked per week by students. A sample of 49 students showed a mean of 24 hours with a standard deviation of 4 hours. Develop a 95 percent confidence interval for the population mean. Co) TF ae x 24 + 1.96(4/V49)= 24 + 1.12 The confidence limit is 22.88 to 25.12.TESTS OF HYPOTHESES Statisticians have developed several tests of hypotheses (also known as the tests of significance) for the purpose of testing of hypotheses which can be classified as: (a) Parametric tests or standard tests of hypotheses; and (b) Non-parametric tests or distribution-free test of hypotheses.TESTS OF HYPOTHESES + Parametric tests usually assume certain properties of the parent population from which we draw samples. + Assumptions like observations come from a normal population, sample size is large, assumptions about the population parameters STICBICHT MAE lee Gemrelle * Non-parametric tests do not depend on any assumption about the parameters of the parent population. * Non parametric test generally are less statistically powerful than the analogous parametric procedure * anonparametric test will require a slightly larger sample size to have the same power as the corresponding parametric test.TESTS OF HYPOTHESES The basic distinction for paramteric versus non-parametric is: If measurement scale is nominal or ordinal then use non- parametric statistics If you are using interyal or ratio scales you use parametric statistics. Parametric tests can be used only when the distribution of data is normal. If a distribution deviates markedly from normality then you take the risk that the statistic will be inaccurate. The safest thing to do is to use an non-parametric statistic.TESTS OF HYPOTHESES Analysis Type Example Pammcige — Nuaparanstre Procedure Procedure ‘Compare means between two Is the mean systolic blood Two-sample t-test. Wilcoxon rank- distinct/independent groups pressure (at baseline) for sum test patients assigned to placebo different from the mean for patients assigned to the treatment group? Compare two quantitative Was there a significant Paired t-test Wilcoxon signed- measurements taken from the change in systolic blood rank test same individual pressure between baseline and the six-month follow- up measurement in the treatment group? Compare means between If our experiment had three Analysis of variance Kruskal-Wallis three or more groups (e.g., placebo, new (ANOVA) test distinct‘independent groups drug #1, new drug #2), we might want to know whether the mean systolic blood pressure at baseline differed among the three groups? Estimate the degree of Is systolic blood pressure association between two associated with the patient's quantitative variables age? Pearson coefficient Spearman’s rank of correlation correlationTESTS OF HYPOTHESES ¢ The important parametric tests are: (OR easton (2) t-test; (3)* x2-test, and (4) F-test. ¢ All these tests are based on the assumption of normality i.e., the source of data is considered to be normally distributedTESTS OF HYPOTHESES * This is a most frequently used test in research studies. * z-test is generally used for comparing the mean of a sample to some hypothesised mean for the population in case of large sample, or when population variance is known. * This test may be used for judging the significance of median, mode, and coefficient of correlation.One sample X—Un, ep /vn Incase Gy is not known, we use G, inits place calculating we n—1 is used when two samples are drawn from the same population. In case Op is not known, we use pp initsplace calculating Sar = My My where D, = (X, — Kz) D; =(X; - Xn) ¢,-mkitm™ wh +z-test illustration ed A sample of 400 male students is found to have a mean height 67.47 inches. Can it be reasonably regarded as a sample from a large population with mean height 67.39 inches and standard deviation 1.30 inches? Test at 5% level of significance. Solution; Taking the null hypothesis that the mean height of the population is equal to 67.39 inches, we can write: Hp: My, = 6739" Hi: |ly, # 67.39" and the given information as Y = 67.4 ", 0, = 130", n= 400. Assuming the population to be normal, we can work out the test statistic z as under: eee ee nase WAT 6pfin —130A/400 0065William Sealy Gosset 1 “unpaired t test’. ‘paired t test’One sample Two samples unpaired paired be D-9o rp Pa hf n-1 with d.f=(n—1) where n= number of with df. =(n, +m — 2)TESTS OF HYPOTHESES —S ELL The specimen of copper wires drawn form a large lot have the following breaking strength (in ke. weight): ov 578, 572, 570, 568. 572, 578, 570, 572, 596, 544 ‘Test (using Student's tstatistic)whether the mean breaking strength of the lot may be taken to be 578 ke. weight (Test at 5 per cent level of significance). - ig Ss Solution: Taking the mull hypothesis that the population mean is equal to hypothesised mean of S78 ke., we can write: Hy: = ty, = 578 ke Aye Ug As the sample size is mall (since n= 10) and the population standard deviation is not known. we shall use t-test assuming normal population and shall work out the test statistic tas under: a t on‘To find ¥ and G, we make the following computations SNe : =) 3 ° a a4 é ¢ = i 2 3 4 s 6 7 s 2 wo re = D(X FP _ 572 - 578 Hence. t= =—1488 1272/10 Degree of freedom = (mn — 1) = (10-1) =9s.005|1.376|1.s53|3.o75|e.316|a2.78 lo.7es|0.578|1 2=0|+ esn|> sas 1a> [p.7a3 [0.543 |s.t00|+ sa3[2 1s2|> 776 lp.727 |o.e20|s 180]: «76 [207s |> 570 [9.703 [o.Se5|s.119|+ a15| tess |2ae= lo.7os|o-s85|2.108|+ 57] lp. 700|0.875|+ 053] + a7>| lp.esr |o.e76|s css] 253 p.ees [o.570|t.o75|+ 350] |p.es2|0.c05|s.076|s.245| lo.es1 |o.c20|s.074].241 [p.eso|o.c85|s.072|* 337] = |o.655|5.582|7.087|= 9 o.ess|o.ces|.cce]® ze |o.cs7|o.e20|1.c04|* 22 5.324.797 [2.074 Ba jo.es7|3.055|+ 28/1 723 [2088 2s [o.see|+.o5s|4 ste|t 708 |2.os5 25 S55 |2.0s8|* ats|9 vos [2056 2s [o.S85|s.00|+ 343 [4.703 [2.048 29 o.ss4|2.055|1 311 |1 e85|2 045 20 [o.5%|s.050|+ s03|1 ena|2 021 50 [9.6a5|+.047|s 2o5|t e76 [2.005 =o [o-545|2.045|2 255|t 673 [2.000 =o [p-ea5|t.043 [1 Sea|t.ce4|s.eo 300 [9.845|7.042|1.250|1.ec0|s.050 o.842|1.030|s 2a2]|s.c45|1.S00 £7 a