Bivariate Investigations.


57 views
Uploaded on:
Description
Bivariate Procedures I Overview. Chi-square testT-testCorrelation. Chi-Square Test. Connections between ostensible variablesTypes:2x2 chi-squareGender by Political Party2x3 chi-squareGender by Dosage (Hi versus Med. Versus Low). Beginning stage: The Crosstab Table. Example:Gender (IV)MalesFemalesDemocrat120Party (DV)Republican102Total1122.
Transcripts
Slide 1

Bivariate Analyses

Slide 2

Bivariate Procedures I Overview Chi-square test T-test Correlation

Slide 3

Chi-Square Test Relationships between ostensible variables Types: 2x2 chi-square Gender by Political Party 2x3 chi-square Gender by Dosage (Hi versus Med. Versus Low)

Slide 4

Starting Point: The Crosstab Table Example: Gender (IV) Males Females Democrat 1 20 Party (DV) Republican 10 2 Total 11 22

Slide 5

Column Percentages Gender (IV) Males Females Democrat 9% 91% Party (DV) Republican 91% 9% Total 100% 100%

Slide 6

Row Percentages Gender (IV) Males Females Total Democrat 5% 95% 100% Party (DV) Republican 83% 17% 100%

Slide 7

Full Crosstab Table Males Females Total Democrat 1 20 21 5% 95% 9% 91% 64% Republican 10 2 12 83% 17% 91% 9% 36% Total 11 22 33 33% 67% 100%

Slide 8

Research Question and Hypothesis Research Question: Is sexual orientation identified with gathering alliance? Speculation: Men are more probable than ladies to be Republicans Null theory: There is no connection amongst sex and gathering

Slide 9

Testing the Hypothesis Eyeballing the table: Seems to be a relationship Is it noteworthy? On the other hand, would it be able to be only a chance finding? Rationale: Is the finding sufficiently distinctive from the invalid? Chi-square answers this inquiry What components would it consider?

Slide 10

Factors Taken into Consideration Factors: 1. Size of the distinction 2. Test size Biased coin illustration Magnitude of distinction: 60% heads versus 99% heads Sample size: 10 flips versus 100 flips versus 1 million flips

Slide 11

Chi-square Chi-Square begins with the frequencies: Compare watched frequencies with frequencies we expect under the invalid theory

Slide 12

What might the Frequencies be if there was No Relationship? Males Females Total Democrat 21 Republican 12 Total 11 22 33

Slide 13

Expected Frequencies (Null) Males Females Total Democrat 7 14 21 Republican 4 8 12 Total 11 22 33

Slide 14

Comparing the Observed and Expected Cell Frequencies Formula:

Slide 15

Calculating the Expected Frequency Simple equation for expected cell frequencies Row all out x segment absolute/Total N 21 x 11/33 = 7 21 x 22/33 = 14 12 x 11/33 = 4 12 x 22/33 = 8

Slide 16

Observed and Expected Cell Frequencies Males Females Total Democrat 1 7 20 14 21 Republican 10 4 2 8 12 Total 11 22 33

Slide 17

Plugging into the Formula O - E Square Square/E Cell A = 1-7 = - 6 36 36/7 = 5.1 Cell B = 20-14 = 6 36 36/14 = 2.6 Cell C = 10-4 = 6 36 36/4 = 9 Cell D = 2-8 = - 6 36 36/8 = 4.5 Sum = 21.2 Chi-square = 21.2

Slide 18

Is the chi-square critical? Importance of the chi-square: Great contrasts amongst watched and anticipated that lead would greater chi-square How huge does it need to be for centrality? Relies on upon the "degrees of flexibility" Formula for degrees of opportunity: (Rows – 1) x (Columns – 1)

Slide 19

Chi-square Degrees of Freedom 2 x 2 chi-square = 1 3 x 3 = ? 4 x 3 = ?

Slide 20

df P = 0.05 P = 0.01 P = 0.001 1 3.84 6.64 10.83 2 5.99 9.21 13.82 3 7.82 11.35 16.27 4 9.49 13.28 18.47 5 11.07 15.09 20.52 6 12.59 16.81 22.46 7 14.07 18.48 24.32 8 15.51 20.09 26.13 9 16.92 21.67 27.88 10 18.31 23.21 29.59 Chi-square Critical Values * If chi-square is > than basic worth, relationship is noteworthy

Slide 21

Chi-Square Computer Printout

Slide 22

Chi-Square Computer Printout

Slide 23

Multiple Chi-square Exact same methodology as 2 variable X 2 Used for more than 2 variables E.g., 2 x 2 x 2 X 2 Gender x Hair shading x eye shading

Slide 24

Multiple chi-square case

Slide 25

Multiple chi-square illustration

Slide 26

The T-test Groups T-test Comparing the method for two ostensible gatherings E.g., Gender and IQ E.g., Experimental versus Control bunch Pairs T-test Comparing the method for two variables Comparing the mean of a variable at two focuses in time

Slide 27

Logic of the T-test A T-test considers three things: 1. The gathering implies 2. The scattering of individual scores around the mean for every gathering (sd) 3. The span of the gatherings

Slide 28

Difference in the Means The more remote separated the methods are: The more certain we are that the two gathering means are diverse Distance between the methods goes in the numerator of the t-test recipe

Slide 29

Why Dispersion Matters Small fluctuations Large changes

Slide 30

Size of the Groups Larger gatherings imply that we are more positive about the gathering implies IQ case: Women: mean = 103 Men: mean = 97 If our example was 5 men and 5 ladies, we are not that sure If our specimen was 5 million men and 5 million ladies, we are significantly more sure

Slide 31

The four t-test formulae 1. Coordinated examples with unequal fluctuations 2. Coordinated examples with equivalent fluctuations 3. Free specimens with unequal changes 4. Autonomous specimens with equivalent fluctuations

Slide 32

All four formulae have the same Numerator X1 - X2 (bunch one mean - bunch two mean) What separates the four formulae is their denominator is "standard blunder of the distinction of the signifies" every equation has an alternate standard mistake

Slide 33

Independent example with unequal differences recipe Standard mistake equation (denominator):

Slide 34

T-test Value Look up the T-esteem in a T-table (use supreme quality ) First decide the degrees of opportunity ex. df = (N1 - 1) + (N2 - 1) 40 + 30 = 70 For 70 df at the .05 level =1.67 ex. 5.91 > 1.67: Reject the invalid (means are distinctive)

Slide 35

Groups t-test printout illustration

Slide 36

Pairs t-test case

Slide 37

Pearson Correlation Coefficient (r ) Characteristics of correlational connections: 1. Quality 2. Noteworthiness 3. Directionality 4. Curvilinearity

Slide 38

Strength of Correlation: Strong, feeble and non-connections Nature of such relations can be seen in disperse charts Scatter outline One variable on x pivot and the other on the y-hub of a diagram Plot every case as per its x and y values

Slide 39

Scatterplot: Strong relationship B O K R E A D I N G Years of Education

Slide 40

Scatterplot: Weak relationship I N C O M E Years of Education

Slide 41

Scatterplot: No relationship S P O R T S I N T E R E S T Years of Education

Slide 42

Strength increments… As the focuses all the more nearly comply with a straight line Drawing the best fitting line between the focuses: "the relapse line" Minimizes the separation of the focuses from the line: "slightest squares" Minimizing the deviations from the line

Slide 43

Significance of the relationship Whether we are sure that a watched relationship is "genuine" or because of chance What is the probability of getting results this way if the invalid theory were valid? Contrast watched comes about with expected under the invalid If under 5% chance, dismiss the invalid theory

Slide 44

Directionality of the relationship Correlational relationship can be sure or negative Positive relationship High scores on variable X are connected with high scores on variable Y Negative relationship High scores on variable X are connected with low scores on variable Y

Slide 45

Positive relationship case B O K R E A D I N G Years of Education

Slide 46

Negative relationship case R A C I A L P R E J U D I C E Years of Education

Slide 47

Curvilinear connections Positive and negative connections are "straight-line" or "straight" connections Relationships can likewise be solid and curvilinear too Points comply with a bended line

Slide 48

Curvilinear relationship illustration F A M I L Y S I Z E SES

Slide 49

Curvilinear connections Linear measurements (e.g. connection coefficient, relapse) can veil a critical curvilinear relationship Correlation coefficient would demonstrate no relationship

Slide 50

Pearson Correlation Coefficient Correlation coefficient Numerical articulation of: Strength and Direction of straight-line relationship Varies amongst –1 and 1

Slide 51

Correlation coefficient results - 1 is an impeccable adverse relationship - .7 is a solid pessimistic relationship - .4 is a moderate antagonistic relationship - .1 is a feeble contrary relationship 0 is no relationship .1 is a frail constructive relationship .4 is a moderate constructive relationship .7 is a solid constructive relationship 1 is an immaculate constructive relationship

Slide 52

Pearson\'s r (connection coefficient) Used for interim or proportion variables Reflects the degree to which cases have comparative z-scores on variables X and Y Positive relationship—z-scores have the same sign Negative relationship—z-scores have the inverse sign

Slide 53

Positive relationship z-scores Person Xz Yz A 1.06 1.11 B .56 .65 C .03 -.01 D -.42 -.55 E -1.23 -1.09

Slide 54

Negative relationship z-scores Person Xz Yz A 1.06 -1.22 B .56 -.51 C .03 -.06 D -.42 .66 E -1.23 1.33

Slide 55

Conceptual equation for Pearson\'s r Multiply every cases z-score Sum the items Divide by N

Slide 56

Significance of Pearson\'s r Pearson\'s r lets us know the quality and course Significance is dictated by changing over the r perfectly proportion and finding it in a t table Null: r = .00 How distinctive is the thing that we see from invalid? Under .05?

Slide 57

Computer Printout

Recommended
View more...