Research Methods and Information Systems

Name

Course

Date
QUESTION 1

The number of Black students in Osgoode’s graduating class is 26. The number of total graduates

is 306. The number of Blacks in the City of Toronto is 9.6% for that same year. Are students who

self-identify as Black underrepresented in Osgoode Law School or is this difference the result of

natural random variation?

a. Write the appropriate hypothesis.

The following are the hypothesis of this study:

Null Hypothesis: The proportion of Black students in Osgoode's graduating class is a true representation

of Black individuals in the City of Toronto (9.6%), p = 0.096

Alternative Hypothesis: The proportion of Black students in Osgoode's graduating class is not a true

representation of the Black individuals in the City of Toronto, p ≠ 0.096

b. Are the necessary assumptions to make inferences satisfied?

The following are the necessary assumptions to make the inferences satisfied:

Random sampling: The sample of graduates should be a representative of the entire graduating class

Large sample size: The number of graduates should be sufficiently large for the one-proportion z-test to

be valid.

Independent observations: Each graduate’s identification as Black should be independent of the others

c. Perform a one-proportion z-test.

To perform the one-proportion z-test, we use the formula: z = (p̂ - p0) / √ (p0 * (1 - p0) / n) where p̂ is

the sample proportion, p0 is the hypothesized proportion under the null hypothesis and n is the sample

size.

The proportion of Black individuals in the City of Toronto is 0.096. Therefore this is the p0. The n is

306(total graduates). The p̂ is 26/306 = 0.08496732. We then substitute these values to get the z.
z = (0.08496732- 0.096) / √ (0.096 * (1 - 0.096) / 306)

z = -0.01103268 / √ 0.0002836078 = -0.01103268 / 0.01684066

z = -0.6551216

d. Connect the p-value to your decision about the null hypothesis.

The z-table was then used to determine the probability corresponding to the area to the left of this value

in the standard normal distribution since the z is negative. The obtained probability is 0.25785 which is >

0.05 alpha significance level. Therefore, we fail to reject the null hypothesis. We accept the null

hypothesis that the proportion of Black students in Osgoode's graduating class is a true representation of

Black individuals in the City of Toronto (9.6%), p = 0.096

e. State an appropriate conclusion

From the analysis, we obtained a p-value of 0.25785, which is greater than the typical significance level

of 0.05, and hence we fail to reject the null hypothesis. Therefore, we conclude that there is no enough

evidence to conclude that the proportion of Black students in Osgoode's graduating class is significantly

different from the proportion of Black individuals in the City of Toronto. This means that the observed

difference may be attributed to natural random variation or factors that may warrant further investigation.

QUESTION 2

From the data provided in the Assignment#3 Data spread sheet, is there sufficient evidence to

claim that hours studying affect assignment grades?

a) State your hypothesis.

The hypothesis for analysis is:

Null Hypothesis: Hours studying do not affect assignment grades, β1 = 0

Alternative Hypothesis: Hours studying affect assignment grades, β1 ≠ 0

b) Are there necessary assumptions to make inferences satisfied?

Linearity: The relationship between hours studying and assignment grades should be linear.

Independence: Each observation should be independent of the others.

Homoscedasticity: The variance of the errors should be constant across all levels of the independent

variable.

Normality: The residual should be normally distributed.

c) Find the regression output for this problem. State the estimated regression equation.

In this analysis, those who did not write the assignment were excluded. A linear regression

model was build to predict the A1 grades using the hours studied and the summary of the results

is shown below (The excel sheet is attached with the assignment showing the summary of the

results). The intercept of the model is 39.4552737. This is the predicted assignment grade when

the number of hours studied is zero. The coefficient of the hours studied is 2.19. This shows that

a unit increase in hours studied leads to an increase in the A1 grade by 2.19 units. The model

produced an R-squared value of 0.409 which means that approximately 40.9% of the variability

in the assignment grades can be explained by the number of hours studied using the linear

regression model. Using the model intercept and the coefficient of the hours studied, the

estimated regression equation is y = 39.4552737+ 2.190921228 * Hours studied

The obtained p-value for the hours studied is 0.046. This is less than the significance level of 0.05

hence we reject the null hypothesis. This means that the number of hours a student studies has a

statistically significant effect on their assignment grades.

In conclusion, based on the regression analysis, we can confidently state that there is a relationship

between hours studied and assignment one grades and the relationship is statistically significant.

QUESTION 3

Assess the evidence in the research published by Liqun Cao (2014). In addition to

providing a short summary of the article, consider the following key questions relating to

validity.

a. What is the dependent variable to be explained and how is it measured? How well does

the measurement represent the intended construct?

Li Cao's (2014) research article "Aboriginal People and Confidence in the Police" explores the

factors affecting Aboriginal people's confidence in the Canadian law enforcement. It looks at

how these perceptions are affected by trust, the local environment, and being a victim of crime.
The study found that, compared to non-Aboriginal Canadians, Aboriginal people had

significantly less trust in the law enforcement. This difference can be attributed to past

experiences of colonial and discrimination, present issues with excessive law enforcement, and

racism. The results highlight the need for actions like cultural sensitivity training, community

engagement initiatives, and increased accountability to overcome systematic biases and build

trust between law enforcement and Aboriginal communities.

The dependent variable in the study is Aboriginal people's confidence in the law enforcement.

This is measured using a 5-point scale ranging from "no confidence at all" to "a great deal of

confidence". The respondents are asked to rate their own degree of confidence hence this

assessment technique directly addresses the concept of confidence in law enforcement

authorities. The measure intends to appropriately capture the target construct, but the validity and

reliability of the measure are not mentioned. Because of this, it is challenging to determine how

effectively the assessment conveys the idea of law enforcement authorities' confidence. Even

though the measurement technique is straightforward and relevant, further information on its

validity and reliability should be acquired to enhance confidence in its accuracy.

b. Is the sample representative of the appropriate population? Are the results

generalizable?

Yes, the sample size for the study consisted of 951 Indigenous (First Nations, Métis, or Inuit)

participants and 21,576 non-Indigenous white participants. The sample is representative of the

appropriate population because a sizable portion of the population is Indigenous, making

comparisons to the non-Indigenous white population relevant. A wide range of opinions and

experiences are ensured by the sample's diversity, which includes individuals from various ethnic

and cultural backgrounds. This enhances the study's generalization to the larger Canadian
population.

The sample used in the research is intended to be representative of Canada's Aboriginal

community, which is a suitable target demographic given the research objective. When

evaluating the information generalizability, it's important to take into account the particular

circumstances of the study, which focused on Aboriginal people in Canada. The sample may

sufficiently represent the characteristics and views of the Canadian Aboriginal community, but it

might not be entirely generalizable to other Aboriginal communities abroad or to non-Aboriginal

populations. The application of the findings to populations outside of the context of Canada may

be influenced by variables like cultural differences, historical circumstances, and regional

variations in relationships with law enforcement.

c. What is the strength of the evidence relating the explanatory or independent variables

to the concept being explained (the dependent variable)?

The strength of the evidence concerning the explanatory or independent variables' relationship to

the concept being explained, confidence in the police among Aboriginal people, is notable. The

study examines the relationship between various explanatory variables, such as age, gender, and

marital status, and Aboriginal people's confidence in the police. The results show that several of

these variables are significantly associated with confidence levels, suggesting a strong

relationship between these factors and confidence in law enforcement agencies. The standardized

coefficients reported unveil significant predictors of confidence in the police among Aboriginal

people. The age, education, marital status, trust, perceptions of crime levels, community

disorder, perceived safety, and police contact as influential factors, all exhibiting statistical

significance (p < 0.05).

d. Are the conclusions and recommendations in the paper fully supported by the
 evidence?

The evidence presented in the research seems to provide strong support for the conclusions and

recommendations made in the paper. Regression analysis offers valuable insights into the factors

that influence Aboriginal people's confidence in the police, identifying important predictors like

age, education, trust, and crime perceptions. These results support the study's conclusions,

showing that demographic factors, trust in the legal system, and perceptions of safety and

disorder all have a significant impact on how this the population feels about the police.

Furthermore, the acknowledgement of the need for additional research indicates a careful

interpretation of the findings and a dedication to continuing research in order to fully understand

the factors influencing the attitudes of Aboriginal people towards the police.
Work Cited

Cao, L. (2014). Aboriginal people and confidence in the police. Canadian Journal of

Criminology and Criminal Justice, 56(5), 499-526. https://doi.org/10.3138/CJCCJ.2013.E05