learningkidunya.com Course: Educational Statistics (8614) Level: B.Ed (1.5 Vears) Semester: Autumn, 2020 ASSIGNMENT No. 2 (Units: 6-9) Q.1__ Define hypothesis testing and logic behind hypothesis testing. fA analyst tests an assumption regarding a d by the analyst depends on the nature of the ANS: What Is Hypothesis Testing? hypothesis by using sample data. Such Hypothesis testing is an act in statistic: population parameter. The methodolo, data used and the reason for the analysis. by using sample data. sis, given the data, ‘ rand sample of the Vv goed NW oviding evidence Statistical analysts test a hypo population being analyzed How Hypothesis T In hypothesis testi on the plausibility Statistical anglSyts of the population. oo nal tes different hypotheses: the The null hy ot c i ers; 2.8, a null hypothesis m: e a alternative i vothes sphaen mean return is g a ‘true. However, one of Four Steps of Hypothesis Tes a of) All hypotheses are tested using a four-step process: The first step is for 0 jaonlyone can be right. The next step is to 5 cb will be evaluated. The third step is to carry out the plan and physically analyze the sample data. The fourth and final step is e jthrer pajaethe null hypothesis, or state that the null hypothesis is pases Os arr Real-World Example of Hypothesis Testing If, for example, a person wants to test that a penny has exactly a 50% chance of landing on heads, the null hypothesis would be that 50% is correct, and the alternative hypothesis would be that 50% is not correct. learwrmimvwhkidumnm y,y=a.corzmlearningkidunya.com Mathematically, the null hypothesis would be represented as Ho: P = 0.5. The alternative hypothesis would be denoted as "Ha" and be identical to the null hypothesis, except with the equal sign struck-through, meaning that it does not equal 50%. A random sample of 100 coin flips is taken, and the null hypothesis is then tested. If it is found that the 100 coin flips were distributed as 40 heads and 60 tails, the analyst would assume that a penny does not have a 50% chance of landing on heads and would reject the null hypothesis and accept the altemative hypothesis. If, on the other hand, there were 48 he; fair and still produce such a result. In ca the analyst states that the difference bet n it is plausible that the coin could be as tHigiwhere the null hypothesis is "accept ythe expegted results (50 heads and 50 tails) and the by chance alone." Partb: The Logic of Hypothes The Logic of Hypothesis Te involves four steps. State the Hypothesis:W. concerns the value Af Define the Decisi 0d: M method involves s Usually we e ti hypothesized valygof If the data are cot NOTE: We do not the opposite hypothe: i heres a big discrepen: thesi hypothesis was wrong. We expand on those steps in this Sq. 05 First Step: State the 5 Stating the ota a gs peels nat the value of a rpaliioa parce TCatiOon” Example: Suppose we have are interested in the effect of prenatal exposure of alcohol on the bin weight of ras, Also, supe LC bad if ws of the population of untreated lab rats is 18 grams- Here are the two opposing hypotheses: The Null Hypothesis (Ho), This hypothesis states that the treatment has no effect, For our example, we formally state: bap that the learningkidunywa.comlearningkidunya.com The null hypothesis (Ho) is that prenatal exposure to alcohol has no effect on the birth weight for the population of lab rats. The birthweight will be equal to 18 grams. This is denoted The Alternative Hypothesis (H1). This hypothesis states that the treatment does have an effect. ‘For our example, we formally state: The alternative hypothesis (HL) is that pren: to alcohol has an effect on the birth weight for the population of lab rats. The birthweight will be different than 18 grams. This is denoted Second Step: Define the Decision Methog We must define a method that lets us d whetherthe sample mean is different from the hypothesized population mean. Th whether (reject null hypothesis) or not (accept null hypo an effect (on birth weight), We will go into details later. Third Step: Gather 2 Now we gather data, We do pm the lation, Example: A ran lea durifig, pregnancy. At birth, we measure the wi f the'Sa table. We ealeulate the a Fourth Step: Mak 5 We make a decision i the'tnéan of the stuff hypothesis about the popgtsjon mean Ifthe data are consistent | hypothesis we éo Colo reasonable, Formally: we do not reject the null hypothesis, If there is a big disc1 the data and the nul! tio we conclude that the null hypothesis was wrot ied aucation Example: We compare the “O34 AG i % ht with the hy esized value, under the null hypothesis, of 18 grams 7308411 Ifa sample of rat pups which Were $45; to prenatal alcohol has a birth weight “near” 18 grams we conclude that the treatement does not have an effect. Formally: We do not reject the null hypothesis that prenatal exposure to alcohol has no effect on the birth weight for the population of lab rats, cons eS a learningegkidunya-.comlearningkidunya.com Tf our sample of rat pups has a birth weight "far" from 18 grams we conclude that the treatment does have an effect. Formally: We reject the null hypothesis that prenatal exposure to alcoho! has no effect on the birth weight for the population of lab rats For this example, we would probably decide that the observed mean birth weight of 13 grams is "different" than the value of 18 grams hypothesized under the null hypothesis. Formally; We reject the null hypothesis that prema sure to alcohol has no effect on the Q2_ Explain types of ANOVA. Des uations in which each type should be used.? ANS: An ANOVA test is a way to find.g rev onexperiment results are significant. In other words, they help you to fi a hypothesis or accept the altemate hypothesis. Basically, you're testing group en them. Examples of when you might want to test differen ‘A group of psychiatrig patients a ere eS: counseling, medication and biofeedback. You want to see etter tha at A manufacturer has tw6 diff e ey Mato know if one see if bfe college IFoo St pcrcal, process is better th Students from di outperforms the other, What Does “ 2 Locman Variance test. One-way has one Two-way has two os imple: brand of cereal, calories. What are “Groups” or “I Groups or levels are different —_ Abe variable. In the above example, your levels for “brand of ce ‘i ight be Luck} Trarms, Raisin Bran, Cornflakes — atotal of three “Bi ee “Calories” CAL. be: sweetened, unsweetened — a total of two levels, Lets ay you ae CALLOD oesins ontineis the most effective treatment for (gag 5 “7: consumption. iT might split the study participants into three group; 20841 pice nsin rn S i G- Medication and counseling, Counseling only, Your dependent variable would be the number of alcoholic beverages consumed per day. learningkidunya.comlearningkidunya.com If your groups or levels have a hierarchical structure (each level has unique subgroups), then use a nested ANOVA for the analysis. What Does “Replication” Mean? It’s whether you are replicating (i.e. duplicating) your test(s) with multiple groups. With a two way ANOVA with replication . you have two groups and individuals within that group are doing more than one thing (i.e. two groups of students from two colleges taking two tests). If you only have one group taking two tests, you would use without replication. Types of Tests. There are two main types: one-way replication. One-way ANOVA between groups: u: difference between them. Two way ANOVA without repli testing that same group. For, take a medication to see if it ay tests can be with or without ant to test two groups to sce if there’s a up and you're double- dividuals before and after they ats of those groups are doing erento trying two WV en lated) groups more than one thing. For exat different therapies. Back to Top \ / One Way ANOVA’ Aone way ANOV, using the F-distribut n at dhe tw sare equal. Therefore, a GgAjficant a wal gr Examples of i ” Situation 1: You(bave a Broup of individ Cn it inta grout; and completing different tasks. Fofgk ul ges of tePEn wes loss and form three groups: green ti Situation 2: Similar Sls are shy thto groups based on an attribute they possess. to weight. You could split pi ng | Boe of people aceording scape overweight and normal) and measure their leg strength on a wi Limitations of the Aone way ‘mova Si cae wpa eguTgcy pon each other. But it won"t tell you whicl Estatistic, you may need to run an ad hoc test (like the Least Significant Difference test) to tell you exactly whieh gowns ditaesua@s3 45-7 508411 Back to Top Two Way ANOVA A Two Way ANOVA is an extension of the One Way ANOVA. With a One Way, you have one independent variable affecting a dependent variable, With a Two Way ANOVA, there are two independents. Use a two way ANOVA when you have one measurement variable (i.e. Be =e 2 wee Becta ay = _ COlearningkidunya.com ‘8 quantitative variable) and two nominal variables. In other words, if your experiment has a quantitative outcome and you have two categorical explanatory variables, a two way ANOVA. is appropriate. For example, you might want to find out if there is an interaction between income and gender for anxiety level at job interviews. The anxiety level is the outcome, or the variable that can be measured, Gender and Income are the two categorical variables. These categorical variables are also the independent variables, which.a i factors can be split into ta kev split into three lit into three levels: . female, and ions of the factors. In this example there The results from a Two Wi d_an interaction effect. The main effect is similar to a C considered separately, What are the interactions among independent variable MANOVA Example Re =e weet ee Bc tech us sey COlearningkidunya.com Suppose vou wanted to find out if'a difference in textbooks affected students” seores in math and science. Improvements mm math and science means that there are two denendent variables, so a MANOVA is appropriate. An ANOVA will give vou a single (univariate) f-value while a MANOVA will give you a multivariate F value. MANOVA tests the multiple dependent variables by creating new artificial, denendent variables that maximize ergup differences. These new denendent variables are linear combinations of the meast i Interpreting the MANOVA results If the multivariate F value indicates tha that something is significant. In the ab improved, science scores have improvet would then have to look at cach indiv y significant, this means would not know if math scores have ce you have a significant result. you ic univariate F tests) to see which Bechara MANOVA is many A three-way ANOVA has three factors (independent vaniables) and one dependent variable. For example, time spent studying, prior knowledge. and hours of sleep are factors that affect how well vou do on a test learninghkidunywa-comtearming Kkichanya-comm within groups. In factorial ANOVA, each level,and factor are paired up with each other (“crossed"). ‘This helns vou to see what interagins are eging on between the levels and factors, If there is an interaction then the diffemncesgiidiag factdy depend on the differences in another Let's sav you were ninning a two-way v ale/female performance on a final Xam. bjects had cither had 4. 6. Murs ep iO gia questions: x amain effect? Lp othe: i, & mind © ol nificagtly on their exam performance? 8 Is sleep a main effelyah othd gv MI He. or pli of sleep differ sunificanily in thely gor .) i | enoWw Is there a signific: fe he di ffercag levegpyt slecy rs of sleep and sex iiteractidh he interact with regardadiexam Agri Can any diffiGos Assumptions QMactafhil Normality: the dn embsdlink Tsirigatlll a Independence: Oifg)varieieantictaups or Wile lon ugha aR cr Equality of Variance® day papery bitaces are cgudl dere ors lhe i hi ¢ Y How to run an ANOW SF > EN LS ; These tests are very time-ethg b ath ates very cgecfy want to use software Didab ic iat P For example, several option cl 2013 == 7-—Jucation Running the test in Exe Re =e wwe ewes Bend oe ey = _ oelearningkidunya.com ou have ordered data. you'll want to run a non-parametric ANOVA (usually found under a different heading in the software, like ‘nonparametric tests“) Steps Ibis unlikely you'll want to do this test by hand but if vou must, these are the steps vou'll want to take: r) Find the mean for each of the sraups.\ f Find the overall mean (the mean of thelg\dpSaepmib ABH. Find the Within Group Variation; the total da#Wigp gffcach member's scare from the Group Mean. b Group Mean from the Overall Mean. kithig Group Variation Find the Between Group Variation: the, Find the F statistic: the ratig of B ANOVA ¥s A Student's t-test will tell y ant van dveen groups. A t-test compares means, while the 4 i ca a oy ~ ppulations. You could technically, pes — bveverggs the groups grow in number, vou may enditp with = = iy. ANOVA will give jou a single pumbel a z cs Em mY) eject the null hypothesis. J Back to Top Repeated 2S 3 A ret SERA Ne a difference: you pt relited@icroups. Wi inde Pe lip Pas) ted Measures becauf)he Stine’ Praup of pAWticiparts is elf BGO onde? again. For example, vou cou) icles eee es ents at 1 3.and 6: months after chan Sey circ | ea oom aa e is “time” and the ey endent variahMaeeyo lester Miple is ue called the within Sa B sures ANOV AT QR) gato 0 githptegnultiyen Best Jn both tests. the sc particinants are —- over and ov€r, However With Ted measures the same s condition. For example. blood pressure is measured pristic that chan: rate over time. Forexumple, you col Reasons to use Repeated — —_— A When vou collect data from time, individual differences a source of between group ditfePEhoes ire ettiniriat Testing is more powerful because the sample size isn’t divided between croups. The test can be economical, as vou're using the same participants. Assumptions for Repeated Measures ANOVA learninvghkidunya.comlearningkidunya.com es ANOVA will be valid only if the following assumptions haven't been violated: There must be one independent variable and one dependent variable. The dependent variable must be a continuous variable, on an interval scale or a ratio scale, The independent variable must be categorical, either on the nominal scale or ordinal scale. Ideally, levels of dependence between pairs of grouns is equal (“sphericity”), Corrections are el." Click “Repeated Measures.”” Grephs Utlibes Add-ons Window Help Soman, + (lth Ml Compare Means , Bunwanare.. Generalized Linear Models | E> wutivaist ges Models . Cometate , Rearession , Logtinear r , . , Classity Pres ® 2 ninble has been 4 weeks. this Step 3: Enter the “* measured. For exami number would be 4. BSE ucation 0345-7308411 learningkidunya.comlearningkidunya.com BS Education 0345-7308411 learningkidunya.comlearningkidunya.com Week_ (1, Training) Wet 212 raining) Week 318 Training) Week_4/4,Training) Step 7: Clie Horizontal / Be =e ewe tes Bc ted us ay 8 CORlearningkidunya.com C for box on the night 10: Click the Howing check boxes: Compare main effects. Descriptive Statistic Back to Top Sphericity In statistics, sphericity (¢) refers to Maud by John W_ Mauchly. who c Definition tv test. which was developed in 1940 electronic computer. the assumption. If ‘avoid increases in BS Education 0345-7308411 learningkidunya.comlearningkidunya.com hrauciny’s Test of Sphericry Hessure: MEASURE | MOH | Ripe criequare | at ase [aT ara Tea | 90g [500] Tests the null hypothesis that the ervor covaianc= mai: of the orthonermalized ‘anstormes degendent yas abies |= propomonal to sn isentry mists @. May baused 10 adjust ne deprnes of nascar tor ine averaged twsts of sigoificance. Corrected tests are laplayed in the Tests of neni Subjects Effects table. be Beslan: Intercept Within Subjects Design! TIME Image: UVM.EDU You soul bs port the pha c relationship. ANS: The correlati between the relative en -1.0:and 1.0. A calculated number greater th yas an error in the correlation measurement. Ao a negative correlation, while a correlation of 1.0 shows a perfect positive correlation. A correlation of 0.0 shows no linear relationship betwee Correlation sa EBA a ERG curtain coefficient could be calculated to determine the Tevel of correlation between the price of crude cil and the stock price of an oil-produci cha py apy omton, Sine oi Sompanies eam prestex profi kcal so: ose arabia highly positive. Understanding the Correlition Coefficient There are several types of correlation coefficients, but the one that is most common is the Pearson correlation (r). This measures the strength and direction of the linear learminmwkidunmnywa-comlearningkidunya.com relationship between two variables. It cannot capture nonlinear relationships between two variables and cannot differentiate between dependent and independent variables. A value of exactly 1.0 means there is a perfect positive relationship between the two variables. For a positive increase in one variable, there is also a positive increase in the second variable. A value of -1.0 means there is a perfect negative relationship between the two variables. This shows that the variables move in opposite directions - for a positive increase in one variable, there is a decrease in the second variable. If ee orrelatiqn between two variables is 0, there is no linear relationship between them. The strength of the relationship varies coefficient, For example, a value of 0.2 variables, but it is weak and likely unit consider correlations important until the n the value of the correlation positive correlation between two ts in some fields of study do not gs.at least 0.8. However, a correlation yesenta very strong relationship. nds in the financial markets, the economy, and st Correlation coefficient values Correlation Statistigs and Inve The correlation bet not gonsidered significant. pa in the financial mutual fund ne ding a low or esto . a gee: ae hedge the price rigfepf a Eases bedi they want the divi ¢ gecur Correlation statistics feawhien the corrélayion between two variables changes. For ex: sitive correlation to interest rates since loan rates ‘interest rates. If the stock price of a bank is falling while inte can glean that something's askew. If the stock eee are alsg rising, investors can conclude that the declining b not du gas nde [y-performing bank is likely dealing with : TOT Correlation Coefficient Equation To calculate the Pearson reo he ai CyB at: determine the covariance of the two variables in questi®f. Next, offe mist tilctfattre: iable's standard deviation. The correlation coefficient is determined by dividing the covariance by the product of the two variables’ standard deviations. \begin faligned} &\rho_{xy} = \frac { \text{Cov} (x.y) } { \sigma_x \sigma_y } \\ &\textbf{where:} \\ &who_{xy} = \text{Pearson product-moment correlation coefficient} \\ Bez ew we twe > Bc tc wey COlearningkidunya.com &\text {Cov} ( x,y )= Mext {covariance of variables } x \text{ and } y\\ &\sigma_x = \text {standard deviation of } x \\ &'sigma_y = \text{standard deviation of } y \\ \end aligned} pxy=oxoyCov(x,y)where:pxy=Pearson product- moment correlation coefficientCov(x.y)-covariance of variables x and yox =standard deviation of xoy=standard deviation of y Standard deviation is a measure of the di of data from its average. Covariance is a measure of how two variables change togeth tits jitude is unbounded, so it is difficult to interpret. By dividing covariance bythe of thé two standard deviations, one can calculate the normalized version of the’ stati! ¢ correlation coefficient. partb: If we wish to label the strength of | regarded as very weak, 0.2-0.39 as we, very strong correlation, but these be considered. mn, for absolute values of r, 0-0.19 is moderate, 0.6-0.79 as strong and 0.8-1 as g.context of the results should Qa papi chi sq h h ‘ it be applied? for one nominal i he data can be ine variable and each between two nom variable is comp: displayed in a cont column repre h a wants to examine the 2 n ai upd Priel flow). The chi-square test 3 it I hypothesis for this test is. fernative hypothesis is that empathy females Calculate Chi Square Stat First we have to calculate th i : ite We can calculate the expected value of the twn nemana Rarittles by Ying WAS formal Lis Bia Bitten 2, =“e=S Education Where Au ~expected value §=—§ QW. 5-7 308411 212% = Sum of the ith column Like = gum of the kth ror N= total number learningkidunywa.comlearningkidunya.com After calculating the expected value, we will apply the following formula to calculate the value of the Chi-Square test of Independence: Bok (Oy- 2B S Sun sud i Fe id 2 = Chi-Square test of Independence ‘us = Observed value of two nomi 41 = Expected value of two nominal Degree of freedom is calculated by usi DF =(r-1X(c-1) Where DF = Degree of freedom r= number of rows c= number of columns variable Hypotheses Null hypothesis: Assumes thi a Altemative hypothesis; Assu i tion be th variables, Hypothesis testing: z i it is for other tests like ANOVA, ate ‘ 5 ical value. The critical value for tbe squa and the degrees of freedom. alculated using the followit number of value null oN 2. t is applied i to determine - Z F fer (male or female) and voting preference (Democrat, R could use a chi-square test for independence to Sciraine whether meter is a to win preference. The sample problem at the end of When to Use Chi-S Catl Q [ l The tet procedure altho in tie Tests Spnrohin where folswving condliion are The sampling method is sin oth nih 08411 The variables under study are each categorical, If sample data are displayed in a contingency table, the expected frequency count for each cell of the table is at least 5. This approach consists of four steps: (1) state the hypotheses, (2) formulate an analysis plan, (3) analyze sample data, and (4) interpret results. when you have two whether there is a signifi We se ee ee ee _learningkidunya.com State the Hypotheses Suppose that Variable A has r levels, and Variable B has ¢ levels. The null hypothesis states that knowing the level of Variable A does not help you predict the level of Variable B. That is, the variables are independent. Ho: Variable A and Variable B are independent. Ha: Variable A and Variable B are not indepe: + The altemative hypothesis is that knoyving the of Vanable B. Note; Support for the alternative hypot relationship is not necessarily causal, in cof Variable ean help you predict the level the variables are related; but the ne variable "causes" the other, Formulate an Analysis Plan The analysis plan describes how to use gcept or reject the null hypothesis. The plan should specify the fol n Significance level. Ofien, rest: | to 0.01, 0.03, oF 0.10; but significant relationship between 6 ‘Analyze Sample Dail Using sample data, find the deg value associated a sample problem at Degrees of ae =(r- 1). hee ris the levels for the other categoric: Expected frequendigh for each level of one categorical variabl iter * c expected frequencies, accordin| Erc=(nr*ne)/n where Er.c is the expected ey level © of Variable B, nr is the total number of sample i i fable A, ne is the total number of sample observations hd level c of Variable B. statistic as extreme as the test llue is the probability of observing a samp! he test statistic is a chi-square ¢ the Chi-Square Distribution Calculator to he probability associated with the test statistic, Use the deerees of freedom computed learrwrminmwvwkidunmyw-a_corlearningkidunya.com Inter pret Resul nm hypothesis. Typically, this involves comparing the P-value to the significance level, and rejecting the null hypothesis when the P-value is less than the significance level, Test Your Understanding, Problem £1000 voters. Respondents were se steps below: ative hypothesis. State the hypothes He ence ay — OL 5 F002 180 ca a 180 400/270 + 900/270 + 100/60 Re-=e went we ss Bend un way = _ oolearningkidunya.com where DF is the derees of freedom. ris the number of levels of gender he number of levels of the voting preference, nr is the number of ob: ions from level r of gender, ne is the number of observations from level ¢ of voting preference. n is the number of observations in the sample. Er.c isthe expected frequency count when gender is level rand voting preference is level c. and Or.c is the observed frequency count when gender is level r voting preference is levelc. The P-value is the probability that a degrees of freedom is more extreme than 16.2. " i We use the Chi-Square Distribution € 7. Ais <2 > 16.2) = 0.0003 nee the P-value (0.008 Hes, ‘the significance level (0.05), we cannot Interpret results, pt the null hypothesis. Thus, we coni ¢is.a relationship between gender and yoting pref Description, Inferential Prediction, Designed Statistics Experiments r 1, R®, R“adjusted Paired 2, 3, or more No differentiation between the 1 or more independent x's, Variables. 4 dependent: y = f(x) implicitly Explicitly: y =a + bx Does not address Attempts to show Reproduced by permission of John Wiley and Sons, Inc. from the book Statistics from A.to Z ~ Confusing Concepts Ciarified ww ew Correlation analysis describes the present or past situation, It uses Sample data to infer a property of the source Population or Process. There is no looking into the future. The purpose of Linear Regression, on the other hand, is to define a Model (a linear equation) which can be used learmin2vzhkidunya.comlearningkidunya.com to predict the results of Designed Experiments. Correlation mainly uses the Correlation Coefficient, r. Regression also uses r, but employs a variety of other Statistics. Correlation analysis and Linear Regression both attempt to determine whether 2 Variables vary in nsyach. Linear Correlation is limited to 2 Vafidbles, which can be plotted on a 2-dimensional 183 orm ables! dimensions. In Correlation, we ask to what degree the, imaginary line that would go through it. Regression, that line is the whole point, afforms a shape that seems to follow an to specify that line. In Linear -fit line through the data: y = a + bx. ed toyguantify the association betwee v i 5 e epenl and a dependent variable or betwee! p iables): si alysis is yélated technique to assess the relation Q eand g ore its or confounding vari e variableds en fo ~— variable and th ris! tors, or explanato ralysi ores is denoted "y" a d i. [NOTE: The te ca ading if ite ability to ain even beyond the limits an impression of i \2 associations. The terms "indepen lable a tof ‘interpretations as they do not strongly imp! Correlation Analysis me In correlation analysis, we estimate a-Yumple correli one ‘ent, more specifically the Pearson Product ion. Wc The ALOU. correlation coefficient, denoted r, ranges between -1 1100... association between the two variables. The athe between two variables can be positive (ie., higher levels of one vas S ah stait BOT] of the other) or negative (ie., higher levels of one vanitbl ial fo ‘of the other). The sign of the correlation coefficient indicates the direction of the association. The magnitude of the correlation coefficient indicates the strength of the association. learningkidunya.comlearningkidunya.com For example. a correlation of r= 0.9 suggests a strong, positive association between two variables, whereas a correlation of r = -0.2 suggest a weak, negative association. A correlation close to zero suggests no linear association between two continuous variables. LISA: [I find this description confusing. You say that the correlation coefficient is a measure of the "strength of association", but if you think about it, isn’t the slope a better measure of association? We use risk ratios and odds ratios t@ quantify the strength of association, ic, when an exposure is present it has how mani ti quantity in correlation is the slope, i.e. many times is the dependent variable is a measure of how much of the variabi differences in the independent variabl a dichotomous outcome wo e be attributed to the presencs fement in the independent variable, how ‘And "r" (or perhaps better R-squared) dent variable can be accounted for by asure for a dichotomous variable and the proportion of ¥ that can It is important to note that there variables, but computation of car dgtect this. Therefore, itis always important t uate the a a correlation coefficient. Graphical displays bgffertis é ng betwen Wariables. The figure below shows four hy along the X-axis ie ‘ oO. on ~ BS Education 0345-7308411 learningkidunya.comlearningkidunya.com 11.Strong, Postive * j 2.Weak, Positive: o * ° 4. Strong, Negative correlation between i Scenario 2 depicts a wi body mass index (which ten might o'See between age and Scenario 3 might depict the lack SPv@focia = 0) Tetween tho extent BE tied exposure in adolescence and age at whic! dolescents initiate sexual activity. Scenmio 4anight de = abietiad betwies theaamber of hours ! Example - Correlation St cecational Age and Birth Weight A small study is conducted i fa sociation between scoatonal gat ah mea NON od Co Lew ordi rane learningkidunya.comlearningkidunya.com > = a BS Education 0345-7308411 learningkidunya.comlearningkidunya.com uuu 3500 ‘Birth Weight, grams 20 0 350 00 50 Gestational Age, weaks » Each point represents an (x,y) pal measured in weeks, and the birth weight, meast ing i z c ble oo ¢ horizontal axis (or X-axis), ad ndent -varial i a ON -axi8), Phe scatter plot shows a positive or dire v c its with shorter gestational aie gestational age; The frm we id ws, =o dy, dal 2X xy 3 ani fe se BSc The variances of x s around their respective sample means ( XandY , considered separa iO) Fidh Gora fe] afiability of the (x.y) pairs around the mean of x and mean of y, considered simultaneously. To compute the sample correlation coefficient, we need to compute the variance of gestational age, the variance of birth weight and also the covariance of gestational age and birth weight. ‘We first summarize the gestational age data. The mean gestational age is: learninzgkidunya-comlearningkidunya.com To compute the variance of gestational age, we need to sum the squared deviations (or differences) between each observed gestational age and the mean gestational age. The cony =a are summarized below. The variance of Pus ‘Next, we summal = © y EY _ 49364 Sie variance of birth wo aa BS Fdncation 0345-7308411 learningkidunya.comlearningkidunya.com 2005 897 804,609 EY 549,334 S(v-YieQ EC-V = 7,767,680 Cov(x. To compute Sos estat ight, deviation from A ec gen far each aie ipant (i.¢., oe ee —\ The is are 5 mein 1 Pom from the mean gestational age and bi ae the table below and multiply, BS Education 0345-7308411 learningkidunya.comlearningkidunya.com Ahr r lation. ‘meaningful be as small as 0.4 ts to determine i.¢., statistically significantly le correlation is suggestive of astatistically signifi ‘Bi Ski a a t ion Kupper and Muller.1 0345-7308411 learningkidunya.com