Get help from the best in academic writing.

Social Popularity and Academic Success essay help site:edu Geology coursework help

CHAPTER 1

THE PROBLEM AND ITS SETTING

A. Background of the Study The phase of growing from a child to being an adolescent, factored by the observation of academic performance, coupled with peer’s appreciation on the individual’s response to the system; continue to be one of the prime topics of discussion, explored in the educational system. It is recommended to review the importance of peer relationships of children and its arguable influence on their psychological and social development. Observing a relational margin between students’ popularity in adolescence and his/her academic success, there is a superlative part of understanding the question “Does popularity influence academic success, within the boundaries of an educational environment?”.

B. Statement of the Problem The purpose of this study is to uncover the level of the impact of social popularity and academic success of Bachelor of Science in Accounting Technology Education students of UM Tagum College. This study aims to answer these questions: 1. What is the level of the impact of social popularity to students in terms of: 2. What is the extent of students’ academic success in terms of: 3. Is there a significant difference on the extent of the impact of social popularity to respondents when grouped according to: 3.1 Gender

3.2 Year level 4. Is there a significant difference on the extent of students’ academic success when respondents are grouped according to: 4.1 Gender 4.2 Year level 5. Is there a significant relationship between the extent of the impact of social popularity and students’ academic success?

C. Hypotheses Based on the abovementioned problems, the researchers have formulated the following hypotheses: 1. There is no significant difference on the extent of the impact of social popularity to respondents when grouped according to gender and year level. 2. There is no significant difference on the extent of students’ academic success when respondents are grouped according to gender and year level. 3. There is no significant relationship between the level of the impact of social popularity and the extent students’ academic success.

D. Theoretical and Conceptual Framework

E. Significance of the study College Students. This study would be able to help college students know the prerogative of being socially popular as well as its effects on their academic success. Moreover, it would give them the idea on what are the possible instances that they may encounter as to attain academic success. Professors. Professors would benefit from this study as they could better apprehend that the varying level of academic success of the students is caused by the said factors of social popularity. Parents.

Future Researchers. Giving emphasis on the academic success of students who are socially popular could give the future researchers a great idea that they could also conduct similar studies with.

F. Definition of Terms

For better understanding of the readers, terms here are conceptually and operationally defined for: Social Popularity. Social popularity is a situation in which someone or something is popular with many people (Macmillan Dictionary, 2009-2013). It is the condition of being liked, admired, or supported by many people. The quality or state of being popular, especially the state of being widely well-liked accepted, or sought after (The Free Dictioanry by Farlex, 2013).

Academic Success. Academic success means that a person has met or surpassed the goals they set academically. It may be they get certain grades, or a certain GPA. It could be they make the Dean’s or President’s list (Ask, 2013).

Python QUESTION

WILL SEND ALL FILES ONCE QUESTION IS Accepted
GoalsIn this homework you will:
Formulate hypotheses and carry out appropriate statistical tests
Compute confidence intervals based on appropriate assumptions
Work with a real dataset (student behavioral data)
What to SubmitFor this assignment you should submit:
(i) The Writeup.docx file with all your answers filled in
(ii) The Writeup.pdf file with all your answers filled in (same file as (i) but converted to a PDF)
(iii) The problem1.py file showing your work for Problem 1
(iv) The problem2.py file showing your work for Problem 2
All your answers to each question should be clear in the writeup. It is recommended to have your code files print the values you are calculating for each question, but this is not required.
BackgroundBefore attempting the homework, please review the notes on sampling and hypothesis testing on the course website. Feel free to copy and modify any of the code we have provided for you here.
Some motivation and helpful information on hypothesis tests with an example:
A hypothesis test takes the sample you collected and compares how its mean looks with respect to the sampling distribution. So the data you collect as your sample, you calculate a single ‘mean’ value (xbar) to test against the ‘population of means’. This is often done because if one has enough data sampled, the mean is distributed as a Gaussian distribution (per the Central Limit Theorem). The good thing is that even if not much data could be collected, you have the option to use a Student’s t test. The beauty of this is that virtually no matter what data you are testing (rainfall, test scores, salaries, errors, etc.), the methods you’re being taught are applicable, regardless of the type of data you collect.
The ‘assumed population mean’ (null hypothesis) is the value you test against. For example, your friend Jake says “I reckon there’s 3.43957792347 cm of rainfall here daily in spring.” If you want to prove him wrong (by running a hypothesis test), you would:
Collect samples of rainfall each day in spring at that location.
Compute their sample mean.
Temporarily assume Jake is correct. (Null Hypothesis H0: mu = 3.43957792347 cm, Alternative Hypothesis H1: mu != 3.43957792347 cm) (!= means we only care whether Jake is wrong, if we wanted to prove there was more rainfall, H1 would instead be mu >= 3.43957792347 cm)
Hold Jake to a standard (alpha = 0.1, 0.05, 0.01, whatever seems appropriate to prove “your sample was rare beyond a doubt.”)
Calculate the test statistic (the z-score or t-score, based on the number of datapoints you have collected or information on the true standard deviation value) from the equation in the lecture slides. As a rule of thumb from the Central Limit Theorem; if there are more than 29 samples or you know the population standard deviation, use a Z-score. If there are less than 30 samples and you don’t know the population standard deviation, use a T-score
Determine the p-value from your test statistic. Note: For each p-value, there corresponds a ‘critical test statistic’ value (or pair of values if a 2 tail test, but you really just care about the side your test point is on for which of the two you’d look at in that case.) so you could just find that value and see if your value passes it in the ‘more rare’ direction.
Compare the p-value to the alpha value you set. If the p-value is smaller than alpha, you have proved Jake wrong (You reject the null hypothesis). If the p-value is larger than alpha, you have failed to prove Jake wrong, but he is not necessarily right because you have only “failed to reject” the null hypothesis.
Confidence IntervalsConfidence intervals are somewhat a flipped perspective compared to hypothesis testing. Hypothesis testing yields claims like “I have rejected that the average wind speed is 20km/hr under significance value alpha=0.05.” Whereas a confidence interval for the same claim might say something like “I am 95% confident that the true average wind speed falls in the range (15, 19) km/hr.” Here, we notice that the value 20 did not fall in the confidence interval, and so if hypothesis tested at alpha = 1 – 0.95 = 0.05 like we had, this confidence interval already shows the result of the test would be to reject H0: mu=20km/hr.
Why do we have both hypothesis testing and confidence intervals then? There’s a number of reasons, but arguably one of the main distinguishing factors is that hypothesis testing is used for making or disproving claims, and confidence intervals don’t have to involve proving or disproving a claim but can instead provide a valuable way of bounding an unknown value to some level of certainty.
Python FunctionsReading .txt files
There are several different file formats for data, including .csv and .json which we will cover later. One of the simplest is storing in text (.txt) files, which is how the data is provided to you in this homework. To get each line of a text file sample.txt stored as a separate element in a list data, you can write:
myFile = open(‘sample.txt’)

data = myFile.readlines()

myFile.close()

Each element of data will be a string. To convert them to floats, we can use a list comprehension:
data = [float(x) for x in data]

Mean and Standard Deviation
While they are relatively easy to write manually, the numpy library in Python has built-in functions for finding the mean and standard deviation of a list. To import it, we can write:
import numpy as np

The mean of data is found as
avg = np.mean(data)

To find the standard deviation, type
sd = np.std(data, ddof=x)

where x is the differential from the number of samples N to determine the degrees of freedom. Typically, we want ddof=1 (which divides by N-1 instead of N) unless we know the population mean (in which case ddof=0).
Standard Normal and Student’s t Distributions
The two distributions you will rely on heavily in this homework are the standard normal (z) and the student’s t distributions.
To import the standard normal distribution, type
from scipy.stats import norm

Then, to find the probability that a value lies below a particular point z_c, type
p = norm.cdf(z_c)

Inversely, to find the point z_c below which the probability is p (i.e., the inverse cdf), type
z_c = norm.ppf(p)

To import the Student’s t distribution, type
from scipy.stats import t

Then, to find the probability that a value lies below a particular point t_c, type
p = t.cdf(t_c, df)

where df is the degrees of freedom for the t distribution.
Inversely, to find the point t_c below which the probability is p (i.e., the inverse cdf), type
t_c = t.ppf(p)

Real-World Example of Hypothesis Testing:There is a branch of research concerned with analyzing data collected about students to design machine learning-based recommendation systems or personalization engines aimed at improving learning outcomes. Several academic conferences, such as the Educational Data Mining conference, and companies in industry, such as Zoomi Inc., have been created to pioneer this mission statement.
A popular thrust of this research is investigating relationships between how a student interacts with the learning content in a course (their behavioral data) and the knowledge that they gain from the course (their performance data); intuitively, higher levels of interaction should translate to more knowledge transfer. For online learning courses (on platforms such as Coursera), one recently proposed measure of interaction is engagement, which translates factors like time spent, length of content, number of clicks, length of annotations, and other application usage information into a single measure between 0 (no engagement) and 1 (maximum engagement). If you are interested in learning more about educational data mining, start with this journal paper.
Very Important Notes for Your Analysis:In this homework, you will work with the engagement data of about 3,000 students who took an online course. We divide them into two groups: those who demonstrated sufficient knowledge of the material after the course (about 1,000), and those who did not (about 2,000). This determination is made based on whether they passed their final exam. Viewing “knowledgeable” and “unknowledgeable” students as two different populations, your task in Problem 1 will be to formulate and test different hypotheses about their engagement levels.
In short, you will be working with two sampled groups, those who passed the final exam (stored in engagement_1.txt) and those who did not (stored in engagement_0.txt). What is stored in these files is the engagement values of the students from those groups. You will be doing hypothesis tests to determine such things as ‘is the typical passing student’s engagement level 0.75?’ and ‘was there a difference in how engaged passing and nonpassing students were?’
Instructions0) Set up your repositoryClick the link on Piazza to set up your repository for HW 5, then clone it.
The repository should contain two files aside from this readme, both of which you will use in Problem 1:
engagement_0.txt, a text file containing the engagement scores of students who did not demonstrate knowledge of the course material.
engagement_1.txt, a text file containing the engagement scores of students who demonstrated knowledge of the course material.
1) Problem 1: Hypothesis TestingThis problem concerns the datasets of student engagement in engagement_0.txt and engagement_1.txt:
Suppose the instructor of the course is convinced that the mean engagement of students who become knowledgeable in the material (i.e., the engagement_1 population) is 0.75. Formulate null and alternative hypotheses for a statistical test that seeks to challenge this belief. What are the null and alternative hypotheses, and what type of test can be used?
Carry out this statistical test using the engagement_1 sample. Report the sample size, the sample mean, the standard error, the standard score, and the p-value. Are the results statistically significant at a level of 0.05? How about 0.10? What (if anything) can we conclude (i.e., what is the interpretation of the result)?
What is the largest standard error for which the test will be significant at a level of 0.05? What is the corresponding minimum sample size? (You may assume that the population variance and mean does not change.)
Suppose the instructor is also convinced that the mean engagement is different between students who become knowledgeable (the engagement_1 population) and those who do not (the engagement_0 population). Formulate null and alternative hypotheses that seek to validate this belief. What are the null and alternative hypotheses, and what type of test can be used?
Carry out this statistical test using the engagement_0 and engagement_1 samples. Report the sample sizes, the sample means, the standard error, the z-score, and the p-value. Are the results significant at levels 0.05 or 0.10? What (if anything) can we conclude (i.e., what is the interpretation of the result)?
2) Problem 2: Confidence IntervalsIn this problem, consider the following dataset of the number of points by which a sports team won in its last 11 games:
[3, -3, 3, 12, 15, -16, 17, 19, 23, -24, 32]
In other words, a 3 means the team won by 3 points, and a -3 means the team lost by 3 points.
Use the sample to construct a 90% confidence interval for the number of points by which the team wins on average. Report whether you will use a z-test or t-test and report the sample mean, the standard error, the standard statistic (t or z value), and the interval. (Think, which distribution should you use here if very few datapoints are available?)
Repeat Q1 for a 95% confidence interval. What is the standard statistic (t or z value) and what is the interval? Is your interval wider or narrower compared to using the 90% confidence interval?
Repeat part 2 if you are told that the population standard deviation is 15.836. (Think, which distribution should you use here now that you have the true population standard deviation?). Report whether you will use a z-test or t-test and the values for the sample mean, standard error, standard statistic, and confidence interval. Is your interval wider or narrower than the interval computed in Q2?
Assume you no longer know the population standard deviation. With what level of confidence can we say that the team is expected to win on average? (Hint: What level of confidence would you get a confidence interval with the lower endpoint being 0?)

Essay Writing at Online Custom Essay

5.0 rating based on 10,001 ratings

Rated 4.9/5
10001 review

Review This Service




Rating: