Scribd
Upload
5 views6 pages

ECMT1010 Wage Analysis Assignment

economics and statistics assessment

Uploaded by

joeymedxna
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views6 pages

ECMT1010 Wage Analysis Assignment

economics and statistics assessment

Uploaded by

joeymedxna
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

University of Sydney SID

5 4 0 7 3 7 3 4 9

Data set #
9

ECMT1010 Assignment
Due: 11.59PM Friday 17 May 2024

1. Show how you arrive at the number of bins. Describe and compare the histograms (≤ 50
words).

Number
Of
Males

Hourly wage rate (males)

Number

Of

Females

Hourly wage rates (female)

By looking at the number of observations in the data set which n= (100), I was able to √𝑛, which meant
√100 = 10, so this meant I was able to determine that 10 bins were needed.

The two histograms show a skewness to the right, where the mean male wage rate is 6.721, whereas the
mean female wage rate was equal to 4.832. Further the male wage rate exhibited a higher frequency at
increased wage levels.
2. Produce a labelled plot. Report the correlation. Comment on the association (≤ 50 words).

Sample correlation (r) = 0.441

Years of education and wage rates show an adequate correlation that is positive, meaning that between the
two variables there is an adequate association. This means that on average the longer you have an education
the higher wage results you will achieve.

3. Show all steps in the list.

𝐻𝑜 : 𝜌=0 𝐻𝑎 : 𝜌0

Where ‘p’ is representing the correlation between years of education and wages.
0.441 √100−2
Test statistic→ = 4.864
√1−0.4412

The p-value equates to 0.0000044 or approximately rounded to three decimal places 0.00.

At the significance level of 5%, the p-value of 0.0000044 < 0.05, which means you would reject the null
hypothesis and accept the alternative hypothesis.

To conclude, the statistically significant evidence shown, shows there is in fact an association between the
wage rates and the years of education received.
4. Define your notation.

𝑦 = β𝑜 + β1 𝑥 + ∈

𝑦 = (response variable) = wage rate of the workers

β𝑜 = The Y-intercept

β1 = The explanatory variable→the regression line slope.

𝑥 = The years of education.

5. Report regression results using this table.

𝑦̂ = -0.974 + 0.530 𝑥

variable coefficient std error t-statistic


p-value
Intercept -0.974 1.456 -0.669 0.505
education 0.538 0.111 4.863 0.0000044
1
Regarding the slope estimate, for every 1 year of education received, the hourly wage rate increases by
0.538.

6. Show all steps in the list.

𝐻𝑜 : β1 = 0

𝐻𝑎 : β1  0

β1 , is equal to the workers hourly wage rate slope.


β 0.538
Test statistic = 𝑆𝐸1 = 0.111 = 4.847

The degrees of freedom= 100-2=98

The p-value = 0.0000024 x 2 = 0.0000048 due to being a two tailed test.

If p < 0.05, rejecting the null hypothesis is the best option. But if p>0.05 the best option is to not reject null
hypothesis.

Due to 0.0000048< 0.05, we can determine that that due to a 5% significance level the null hypothesis can be
rejected. This means the alternative hypothesis can be accepted. This means we have statistically significant
evidence to reveal that the hourly wage rate is determined by the number of years received in education.

7. Define notation and state hypotheses.

𝜇𝑀 = the average hourly male wage rate in 1978

𝜇𝐹 = The average hourly wage rate for females in 1978

𝐻𝑜 = 𝜇𝑀 - 𝜇𝐹 = 0

𝐻𝑎 = 𝜇𝑀 > 𝜇𝐹

8. Keep your comments brief (≤ 50 words).

The randomised distribution is precisely symmetrical, and bell shaped. The difference between the mean
female and male wage rates is centred at the mean sample of 1.886.
9. Show your steps, test results, and conclusion.

The population mean is estimated by the mean value by looking at question 8 (listed above) by looking at the
bootstrap dot plot which was 1.886. Using a cut off point for the randomised dot plot the p-value we get is
0.0012.

We reject the Null hypothesis as 0.0010 < 0.05.

Through using the Central limit theorem, we can reject the Null hypothesis due the p-value being 0.029,
which 0.030 < 0.05. This conducts that on average, the female hourly rate is less than the male hourly wage
rate.
10. Keep your answer brief (≤ 50 words).

By examining the central limit theorem along with the randomised distribution results, we can determine
that the alternate hypothesis is infact true, this means the hourly wage rate of women being lower than the
hourly wage rate of males in 1978. However, we are unable to determine whether it is correlation or
causality due as the study is an observational and not an experimental.

Common questions

Powered by AI

The regression results show that for each additional year of education, the hourly wage rate increases by 0.538. This slope estimate is statistically significant at a p-value of 0.0000048, which is well below the 0.05 significance level, indicating a strong influence of education on wage determination .

The test statistic is used to assess the significance of the correlation and regression results, with the statistic for the slope being 4.847 and a corresponding p-value of 0.0000048. This is crucial for verifying the role of education in determining wages, ensuring that the observed relationship is not due to random chance .

The dataset shows a positive correlation of 0.441 between years of education and wage rates, indicating that longer education is generally associated with higher wages. The statistical test confirms this association, allowing us to reject the null hypothesis in favor of the alternative, suggesting a significant relationship between education length and wage outcomes .

The p-value of 0.0010 from the test suggests that the wage rate difference between genders was statistically significant, with males earning more on average than females. This indicates substantial evidence against the hypothesis that there is no wage difference between the genders .

The bootstrap dot plot provides a visual representation of the expected distribution of mean wage differences, centering around a sample mean of 1.886. This supports the observed wage disparity, reinforcing the conclusion that male wages were significantly higher than female wages, underscored by a p-value < 0.05 .

The statistical evidence, with a p-value of 0.0010, is robust enough to reject the null hypothesis and accept the alternative that male wages are higher than female wages in 1978. This signifies a gender wage gap and highlights the need for further investigation into its causes and potential policy interventions .

Rejecting the null hypothesis supports the claim that there is a statistically significant positive correlation between years of education and wage rates. This implies that education significantly influences wage determination, confirming that wage outcomes improve with more education, thus affecting economic policy and workforce education strategies .

The square root of the number of observations (√n) is used to determine the number of bins for creating a histogram. With 100 observations, this results in 10 bins. The histograms reveal a right skewness, where the mean male wage rate is 6.721 and the mean female wage rate is 4.832, indicating a higher frequency of higher wage levels for males .

The findings that wage rates correlate positively with education years imply that economic policies should focus on improving educational access and quality to enhance wage outcomes. Such policies could address wage inequities and increase national income levels by investing in education as a strategic economic development tool .

A hypothesis test comparing average male and female wage rates using the randomised distribution and central limit theorem is employed. With a p-value of 0.0012 and the conclusion that 0.0010 < 0.05, the null hypothesis is rejected, supporting the conclusion that the average female hourly rate is less than that of males .

You might also like