Ladies Professional Golf Association Winnings

(LPGA)

VS

Senior Professional Golf Association Winnings

(SPGA)

2-Sample t-Test

TOUR WINNINGS LPGA 1591959 LPGA 1337253 LPGA 956926 LPGA 863816 LPGA 757844 LPGA 679929 LPGA 663356 LPGA 584246 LPGA 583796 LPGA 577875 LPGA 572940 LPGA 538054 LPGA 512273 LPGA 501798 LPGA 497640 LPGA 484759 LPGA 447903 LPGA 440498 LPGA 410973 LPGA 405142 LPGA 370162 LPGA 369176 LPGA 367258 LPGA 355989 LPGA 354131 LPGA 322308 LPGA 303929 LPGA 301086 LPGA 297973 LPGA 296347

TOUR EARNINGS Senior 2515705 Senior 2025232 Senior 1911640 Senior 1513524 Senior 1493282 Senior 1327658 Senior 1167176 Senior 1118377 Senior 1108245 Senior 1087284 Senior 1051357 Senior 1039334 Senior 997318 Senior 993291 Senior 988778 Senior 951072 Senior 882532 Senior 869839 Senior 857746 Senior 816342 Senior 754046 Senior 743841 Senior 737860 Senior 726674 Senior 715035 Senior 710749 Senior 683314 Senior 638621 Senior 635095 Senior 631046

winnings (thousands)0 500 1000 1500 2000 2500 3000

Golfers99 Box Plot

0

48

1216

LPGA

48

1216

Senior

winnings (thousands)0 500 1000 1500 2000 2500 3000

Golfers99 Histogram

MIN Q1 MED Q3 MAX

LPGA 296347 367258 491200 584246 1591959

SPGA 631046 737860 969925 1118377 2515705

MEAN SD

LPGA 558245 299096

SPGA 1056400 446668

We can see from the graph and the relationships between the mean and median for each data set (mean>median), that both data sets are skewed right. SPGA has a larger center than LPGA. This can be seen from the graphs an comparing the median and mean for each data set. SPGA has a larger spread than LPGA. Both data set have outliers on the high end of the data.

Ho :μL =μS(Average LPGA Winning equals average SPGA winnings)Ho : μL < μS(Average LPGA Winning equals average SPGA winnings)

Even Though Both graphs are skewed right combined sample size is equal to 60. This was not a random sample since I gathered the top 30 for each tour so this might cause a problem with conclusion.xL =558,245 xS =1,056,400 xL −xS =558,245 −1,056,400 =−498,155

t=−5.0757 p=0.0000278 df =50.653I will Reject Ho because p < 0.05 and | t | >1.675. My data is statistically significant and I am able to conclude that average winnings on LPGA tour are less than average winning for SPGA tour. Since this was a Reject Ho conclusion it is possible that we have committed a Type I Error which would be concluding that the average LPGA winnings is less than SPGA, when in reality the average LPGA winnings are not less than SPGA.

Ho :μL =μS(Average LPGA Winning equals average SPGA winnings)Ho : μL < μS(Average LPGA Winning equals average SPGA winnings)

The 90% confidence interval for the difference in average winnings(LPGA - SPGA) is (-66000,-33000). Since all the values are negative we can conclude that average winning for LPGA is less than average wining for SPGA

Husband’s Age

VS

Wife’s Ages

Matched Pair t-Test

Husband Wife Husband-Wife 22 21 1 38 42 -4 31 35 -4 42 24 18 23 21 2 55 53 2 24 23 1 41 40 1 26 24 2 24 23 1 19 19 0 42 38 4 34 32 2 31 36 -5 45 38 7 33 27 6 54 47 7 20 18 2 43 39 4 24 23 1 40 46 -6 26 25 1 29 27 2 32 39 -7 36 35 1 68 52 16 19 16 3 52 39 13 24 22 2 22 23 -1 29 30 -1 54 44 10 35 36 -1 22 21 1

Husband Wife Husband-Wife 44 44 0 33 37 -4 21 20 1 31 23 8 21 22 -1 35 42 -7 23 22 1 51 47 4 38 33 5 30 27 3 36 27 9 50 55 -5 24 21 3 27 34 -7 22 20 2 29 28 1 36 34 2 22 26 -4 32 32 0 51 39 12 28 24 4 66 53 13 20 21 -1 29 26 3 25 20 5 54 51 3 31 33 -2 23 21 2 25 25 0 27 25 2 24 24 0 62 60 2 35 22 13 26 27 -1

Husband Wife Husband-Wife 24 23 1 37 36 1 22 20 2 24 27 -3 27 21 6 23 22 1 31 30 1 32 37 -5 23 21 2 41 34 7 71 73 -2 26 33 -7 24 25 -1 25 24 1 46 37 9 24 23 1 18 20 -2 26 27 -1 25 22 3 29 24 5 34 39 -5 26 18 8 51 50 1 21 20 1 23 23 0 26 24 2 20 22 -2 25 32 -7 32 31 1 48 43 5 54 47 7 60 45 15

Husband_Wife-10 -5 0 5 10 15 20

MarriageAge100 Box Plot

5

10

15

20

25

30

35

40

Husband_Wife-10 -5 0 5 10 15 20 25

MarriageAge100 Histogram

MIN Q1 MED Q3 MAX

Husband-Wife -7 -1 1 4 18

MEAN SD

Husband-Wife 1.92 5.04661

We can see from the graph and the relationships between the mean and median for each data set (mean>median), that the data sets is skewed right. We can also see that there are outliers on the high end of the data set.

Ho :μH−W =0(Average Husband's age and average Wife's age are equal)Ho : μH−W > 0(Average Husband's age is greater than average Wife's age)

Even though the graph is slightly skewed right tah sample size was 100. The data is a SRS

xH−W =1.92 t=3.805 p=0.000123 df =99

I will Reject Ho because p < 0.05 and | t | >1.66. My data is statistically significant and I am able to conclude that average Husband’s age is greater than average Husband’s are older than their Wife’s age

Since this was a Reject Ho conclusion it is possible that we have committed a Type I Error which would be concluding that the average Husband age is greater than the average age of their Wife’s, when in reality the average age of a Husband is not greater than the average age of their Wife’s

The 90% confidence interval for the average difference in ages(Husband-Wife) is (1.0821,2.7579). Since all the values are positive we can conclude that average Husband’s age is greater than the average Wife’s age

Ho :μH−W =0(Average Husband's age and average Wife's age are equal)Ho : μH−W > 0(Average Husband's age is greater than average Wife's age)

Football Injuries

VS

Baseball Injuries

2-Propotion z-Test

Sport Injuries Participants Proportion

Football 334420 20100000 0.01664

Baseball 326714 30400000 0.01075

Football

2%

98%

Injuries

Participants

Baseball

1%

99%

Injuries

Participants

H o :PF =PB(Equal Propotion of Football and Baseball Players get injured)Ho :PF > PB(A greater propootion Football players get injured)

p̂F =0.01664 p̂B =0.01075 p̂F −p̂B =0.01664 −0.01074 =0.0059

z=180.266 p=0

I will Reject Ho because p < 0.05 and z >1.645. My data is statistically significant and I am able to conclude that a higher proportion of Football players get injuredSince this was a Reject Ho conclusion it is possible that we have committed a Type I Error which would be concluding a higher proportion of Football players get injured when in reality it is not true that a higher proportion of Football players get injured

It is reasonable to assume SRS

20100000(0.01664)>10 20100000(1-0.01664)>10

30400000(0.01075)>10 30400000(1-0.01075)>10

The 90% confidence interval for the difference in proportion of injuries (Football - Baseball) is (0.00583,0.00595). Since all the values are positive we can conclude that proportion of Football players who get injured is greater than the proportion of Baseball players who get injured

Ho :PF =PB(Equal Propotion of Football and Baseball Players get injured)Ho :PF > PB(A greater propootion Football players get injured)

Spending Money For Space Exploration

VS

Political Perspective

Chi-Squared Test for Independence

Conservative Liberal Moderate

Just Right

212 164 214

Too Little

36 50 47

Too Much

176 162 174

212

164

214

3650 47

176162

174

0

50

100

150

200

250

Conservative Liberal Moderate

Just Right

Too Little

Too Much

Moderate

49%

11%

40%Just Right

Too Little

Too Much

Liberal

44%

13%

43% Just Right

Too Little

Too Much

Conservative

50%

8%

42%Just Right

Too Little

Too Much

212

36

176164

50

162

214

47

174

0

50

100

150

200

250

Just Right Too Little Too Much

Conservative

Liberal

Moderate

Too Much

34%

32%

34%

Conservative

Liberal

Moderate

Too Little

27%

38%

35%

Conservative

Liberal

Moderate

Just Right

36%

28%

36%

Conservative

Liberal

Moderate

Ho : There is no relationshipe between spening beliefs and political perspective

Ho : There is a relationshipe between spening beliefs and political perspective

Conservative Liberal Moderate

Just Right

202.6 179.6 207.8

Too Little

45.7 40.5 46.8

Too Much

175.8 155.9 180.3

It is reasonable to assume a SRS and all expected counts are greater than 5

χ 2 = 6.724 p = 0.15 df = 4

I will Fail to Reject Ho because p > 0.05 and < 9.49. My data is not statistically significant and I am unable to conclude that there is a relationship between spending perspective and political perspective

Since this was a Fail to Reject Ho conclusion it is possible that we have committed a Type II Error which would be concluding that there is no relationship between spending perspective and political perspective when in fact a relationship exist

χ 2

Per Capita Income

VS

Governor's Salary

Linear Regression t-Test

PerCapitaIncome GovernorSalary 20842 87643 25305 81648 22364 75000 19585 60000 26570 131000 27051 70000 36263 78000 29022 107000 25255 110962 24061 115939 26034 94780 20478 75000 28202 130261 23604 77200 23102 104352 24379 85225 20657 95526 20680 95000 22078 70000 28969 120000 31524 90000 25560 127300 26797 114506 18272 83160 24001 112755

PerCapitaIncome GovernorSalary 20046 78246 23803 65000 26791 90000 28047 90547 32654 85000 19587 90000 30752 130000 23345 107132 20271 75372 24661 115762 20556 101140 24393 88300 26058 105035 25760 69900 20755 106078 21447 82271 23018 85000 23656 115345 20432 90700 23401 105402 26438 124855 26718 121000 18957 90000 24475 115899 22648 95000

60

70

80

90

100

110

120

130

140

PerCapitaIncome (thousands)18 20 22 24 26 28 30 32 34 36 38

GovernorSalary = 1.46PerCapitaIncome + 6.1e+04; r2 = 0.084

-40-20

02040

18 20 22 24 26 28 30 32 34 36 38PerCapitaIncome (thousands)

govsal98 Scatter Plot

H o :β =0(No realtionship between Per Capita Income and Governer's Salary)Ho : β > 0(Positive realtionship between Per Capita Income and Governer's Salary)

Te scatter plot and r2 value indicate a weak linear fit. The residual plot shows an unequal spread of the residual values. These fact could result in incorrect results with test.

b =0.05778 t=2.101 p=0.02 df =48

I will Reject Ho because p < 0.05 and t >1.677. My data is statistically significant and I am able to conclude there is a linear relationship between Per Capita Income and Governor's Salary

Since this was a Reject Ho conclusion it is possible that we have committed a Type I Error which would be concluding that there is a linear relationship between Per Capita Income and Governor's Salary when no relationship exist

The 90% confidence interval for the slope of the line of best fit for Per Capita Income vs. Governor's Salary is (0.2936,2.62032) Since all the values are positive we can conclude that there is a positive relationship between Per Capita Income vs. Governor's Salary

H o :β =0(No realtionship between Per Capita Income and Governer's Salary)Ho : β > 0(Positive realtionship between Per Capita Income and Governer's Salary)