Hypothesis Test for Comparing Two Proportions

Hypothesis Test for Comparing Two Proportions In this article we will go through the steps necessary to perform a hypothesis test, or test of significance, for the difference of two population proportions.  This allows us to compare two unknown proportions and infer if they are not equal to each other or if one is greater than another. Hypothesis Test Overview and Background Before we go into the specifics of our hypothesis test, we will look at the framework of hypothesis tests.  In a test of significance we attempt to show that a statement concerning the value of a population  parameter (or sometimes the nature of the population itself) is likely to be true.   We amass evidence for this statement by conducting a statistical sample.  We calculate a statistic from this sample.  The value of this statistic is what we use to determine the truth of the original statement.  This process contains uncertainty, however we are able to quantify this uncertainty The overall process for a hypothesis test is given by the list below: Make sure that the conditions that are necessary for our test are satisfied.Clearly state the null and alternative hypotheses. The alternative hypothesis may involve a one-sided or a two-sided test.  We should also determine the level of significance, which will be denoted by the Greek letter alpha.Calculate the test statistic. The type of statistic that we use depends upon the particular test that we are conducting.  The calculation relies upon our statistical sample.  Calculate the p-value. The test statistic can be translated into a p-value.  A p-value is the probability of chance alone producing the value of our test statistic under the assumption that the null hypothesis is true. The overall rule is that the smaller the p-value, the greater the evidence against the null hypothesis.Draw a conclusion. Finally we use the value of alpha that was already selected as a threshold value.  The decision rule is that If the p-value is less than or equal to alpha, then we reject t he null hypothesis. Otherwise we fail to reject the null hypothesis. Now that we have seen the framework for a hypothesis test, we will see the specifics for a hypothesis test for the difference of two population proportions.   The Conditions A hypothesis test for the difference of two population proportions requires that the following conditions are met:   We have two simple random samples from large populations.  Here large means that the population is at least 20 times larger than the size of the sample. The sample sizes will be denoted by n1 and n2.The individuals in our samples have been chosen independently of one another.  The populations themselves must also be independent.There are at least 10 successes and 10 failures in both of our samples. As long as these conditions have been satisfied, we can continue with our hypothesis test. The Null and Alternative Hypotheses Now we need to consider the hypotheses for our test of significance.  The null hypothesis is our statement of no effect.  In this particular type of hypothesis test our null hypothesis is that there is no difference between the two population proportions.  We can write this as H0: p1 p2. The alternative hypothesis is one of three possibilities, depending upon the specifics of what we are testing for:   Ha:  p1 is greater than p2. This is a one-tailed or one-sided test.Ha: p1 is less than p2. This is also one-sided test.Ha: p1 is not equal to p2. This is a two-tailed or two-sided test. As always, in order to be cautious, we should use the two-sided alternative hypothesis if we do not have a direction in mind before we obtain our sample.  The reason for doing this is that it is harder to reject the null hypothesis with a two-sided test. The three hypotheses can be rewritten by stating how p1 - p2 is related to the value zero.  To be more specific, the null hypothesis would become H0:p1 - p2 0.  The potential alternative hypotheses would be written as: Ha:  p1 - p2   0 is equivalent to the statement p1 is greater than p2.Ha:  p1 - p2  Ã‚  0 is equivalent to the statement p1 is less than p2.Ha:  p1 - p2  Ã‚  Ã¢â€°   0 is equivalent to  the statement p1 is not equal to p2. This equivalent formulation actually shows us a little bit more of what is happening behind the scenes.  What we are doing in this hypothesis test is turning the two parameters p1 and p2  into the single parameter p1 - p2.  We then test this new parameter against the value zero.   The Test Statistic The formula for the test statistic is given in the image above.  Ã‚  An explanation of each of the terms follows: The sample from the first population has size n1.  The number of successes from this sample (which is not directly seen in the formula above) is k1. The sample from the second population has size n2.  The number of successes from this sample is k2.The sample proportions are  p1-hat k1 / n1  and p2-hat   k2 / n2 .We then combine or pool the successes from both of these samples and obtain:  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚  Ã‚   p-hat ( k1 k2) / ( n1 n2). As always, be careful with order of operations when calculating.  Everything underneath the radical must be calculated before taking  the square root. The P-Value The next step is to calculate the p-value that corresponds to our test statistic. We use a standard normal distribution for our statistic and consult a table of values or use statistical software.   The details of our p-value calculation depend upon the alternative hypothesis we are using: For Ha: p1 - p2   0, we calculate the proportion of the normal distribution that is greater than Z.For Ha: p1 - p2  Ã‚  0, we calculate the proportion of the normal distribution that is less than Z.For Ha: p1 - p2  Ã‚  Ã¢â€°   0, we calculate the proportion of the normal distribution that is greater than |Z|, the absolute value of Z.  After this, to account for the fact that we have a two-tailed test, we double the proportion.   Decision Rule Now we make a decision on whether to reject the null hypothesis (and thereby accept the alternative), or to fail to reject the null hypothesis.  We make this decision by comparing our p-value to the level of significance alpha. If the p-value is less than or equal to alpha, then we reject the null hypothesis. This means that we have a statistically significant result and that we are going to accept the alternative hypothesis.If the p-value is greater than alpha, then we fail to reject the null hypothesis. This does not prove that the null hypothesis is true. Instead it means that we did not obtain convincing enough evidence to reject the null hypothesis.   Special Note The confidence interval for the difference of two population proportions does not pool the successes, whereas the hypothesis test does.  The reason for this is that our null hypothesis assumes that p1 - p2 0.  The confidence interval does not assume this.  Some statisticians do not pool the successes for this hypothesis test, and instead use a slightly modified version of the above test statistic.

Life in the Medieval Period essays

Life in the Medieval Period essays The world in Medieval Times could be illustrated as gory, barbaric, and uncivilized; however, some of the greatest advancements in culture as we know it were produced from the mindsets in these stereotyped Dark Ages. While corruption and cruelty was indeed happening in the courts, the church, and the streets, this period in time produced new knowledge and ways of thinking: ways we have adapted today. Europe suffered in this era in more ways than one. While the rich dined in excess, the peasantry did without. People were illiterate, suffering the plagues, or famishing in their miniscule and cramped huts. The life-span for a woman was about 24 years (pg.55). When Rome fell to the Huns, Catholicism fell on the masses: everyone was imposed to be devout Christians. In the minds of the people The Church was indivisible, the afterlife a certainty; all knowledge was already known (pg.27). People were strong in faith until a series of events shattered (pg.29) its credibility. The Church was first to be questioned. Humanism was spreading quickly by way of professors, through the countries and common people were becoming literate. Money was being paid to absolve sins and now, members of the Church were becoming non-celibate; breaking a rule made by the Vicars of Christ. Was this not a sin? Well as it was, only strong families of great power, such as the Medicis or the Aragons, dared to question the clergy (pg.42). Morals were quickly being lost as prostitution and adultery became frequent and it is said to be due in large part to the religious transformation to come. Martin Luther, a name synonymous with revolution, was a strict Catholic and professor until he saw indulgences being traded for sins and this possibly was reason to start his movement to risk his life and create 95 theses challenging the Church. Luther was sought after over and over for heresy but his ide ...

It depends on the question that the writer choose Essay - 1

It depends on the question that the writer choose - Essay Example While the previous era of globalization came up around falling transport costs, the current globalization is linked to the decrease in the costs of telecommunications, which make it easier to propel the world higher and make it a global village 1. These technologies have made developing nations to become big producers other than just relying on the trading of their raw materials to the west and get finished products in return. All these developments showcase a world full of selective understanding in which the main market players will be more coalitions of conjuncture and the traditional alliances built around a strategic interest point. In essence, globalization has led to the creation of more opportunities than harm. The benefits do outweigh the harms largely. The essay seeks to expound on how globalization has affected the word at positive level 2. The concept of globalization has remained a topic of interest, with great debaters focusing much on it. Its proponents have marked it as an exquisite phenomenon to happen in the world while its opponents criticize it strongly as a component linked to create harm rather than improving the existing conditions. One thing is well notable: that despite the challenges globalization faces, there is a great impact it has played on the lives of millions of people in the world3. Those against globalization may present diverse negative effects of its existence, such as environmental degradation, inequality in economic patterns, loss of jobs and other catastrophic events. However, a good number of developing countries are benefitting immensely from the vice more than before, in terms of economic growth, sustenance of culture and other angles. First, globalization has created many employment opportunities especially in developing nations. Many companies are reviving their operations especially with reference to ICT sector. There

Contemporary Risk Management - Climate Change Essay

Contemporary Risk Management - Climate Change - Essay Example Many of these issues have direct implications for the risk management and insurance industry. Global warming refers to the gradual warming of the earth’s atmosphere that has been documented over the last 150 years. Global warming can result from natural fluctuations in climate or can be induced by human causes. There had been considerable debate as to whether human activity in producing increased emissions and concentrations of greenhouse gases has been associated with the global warming trend. Greenhouse gases effectively trap the heat in the atmosphere. Greenhouse gases, principally CO2, but also ozone, methane, and nitrous oxide, are produced by burning fossil fuels like oil, coal, and natural gas. Energy producers, automobile users, manufacturers, and households all burn fossil fuels and produce greenhouse gas emissions. The rising concentrations of CO2 that are causing global warming principally come from energy production, factories, homes, automobiles, and deforestation. Rising CO2 concentrations are a symptom of expanding economies, jobs, and standards of living. Indeed these economic benefits form the most frequently sited argument for resisting the Kyoto Treaty and other control techniques. For instance, President Bush stated in rejecting the Kyoto Treaty that it would â€Å"have a negative economic impact, with layoffs of workers and price increases for consumers† (Sanger, 2005, 24). Some studies have indicated that agriculture and forestry will actually improve from global warming. For instance, the NAST report finds that U.S. crop productivity is very likely to increase over the next few decades because of global warming. While benefiting consumers, falling crop prices and competitive pressures are likely to stress some farmers. The NAST report also found that forest productivity is likely to increase over the next

What are mitigating factors in learning math in saudi arabia Assignment

What are mitigating factors in learning math in saudi arabia - Assignment Example The discussion of the results basically focused on elaboration of the behavioral patterns of students in relation to motivation in undertaking math as a subject. MRP6: This was the last section of the paper; it focused on a discussion of the purpose of the research followed by the conclusion section that basically provided summarized information of the entire research. Adler, N. E., & Stewart, J. 2009: Recommended reading for the research; provides pertinent information regarding motivation among human beings. This reference has noted that many human beings have a natural tendency of avoiding work. The DOS Based VBPro: This is the content analysis software that was used to facilitate the analysis of the data collected in this research. The rational for selecting this software for the research is based on the fact that it is easily applicable in analyzing qualitative data. Manipulative in learning: This is one factor that arised in the course of data analysis; manipulative method as a form of learning entails a scenario whereby student utilizes certain items i.e. calculators, toothpicks and markers to support hand-in learning. In my research, this is one factor that appeared to among the motivating factors used among students to learn math. School Quiz Logic Puzzles: This is another issue that emerged as I analyzed my data. It is a tool that was applied in most Saudi Schools to improve the performance of students in mathematics. Logic puzzles are basically brain games where a student is subjected to a scenario where he has to apply logic in order to solve a problem. Qualitative Techniques: This is the procedures that were used to collect and analyze data that guided my research. The rationale for selecting qualitative technique as my research designed was based on the fact that it facilitated asking and answering questions such as: Why? Where? How?

Critically discuss investor sentiment and the implications of their Essay

Critically discuss investor sentiment and the implications of their behaviour - Essay Example The role of investor sentiments and how it can affect their behavior has been in relatively direct contrast with the modern portfolio theory. Modern portfolio theory indicates that the investors always act rationally and also take into consideration all of the available information. However, a large number of empirical studies have actually shown the irrational behavior of the investors as well as repeated errors in overall judgment. The behavioral finance as a field therefore focuses upon understanding as to how such cognitive behaviors can be explained besides exploring as to why such errors occur in investor judgments. Behavioral finance therefore uses the theories from psychology as well as sociology and other disciplines to actually explore and understand basic investor behavior and how it may have an impact on the market. Investor Sentiments are generally defined as beliefs in the future cash flows as well as the investment risks which cannot be otherwise defined by the facts at hand. This is based upon the assumption that the investors are actually subject to sentiments and their decisions are subject to the way their sentiments interact with their overall decision making process. What is also critical to note that betting for such sentiments however, can have a relatively high risk? As such there are fundamental trade-offs need to be made in balancing the role of sentiments and the risks taken based upon those sentiments.( Ackert & Deaves, 2010). There have been many episodes where the investor sentiments actually drove the prices up without the fundamentals of the company or market supporting the same. The internet bubble as well as the inflated prices of telecom stocks on NASDAQ indicates that the investor sentiments can actually drive the prices to higher levels without actually assessing the actual risks and rewards associated with particular

The Third type of failure in a distributed system Essay Example for Free

The Third type of failure in a distributed system Essay In this paper I will be discussing the issue of failures in a distributed system, and to understand the different failures I will write about four failures that occur in and affect a distributed system. Also, I will be discussing and writing about how to isolate and fix two out of the four failure that can occur in the distributed system. In a distributed system nothing is set in stone or perfect, so there are some issues that can arise, and the issues that arise are the failures that can occur in these distributed systems. The failure that can occur are Fail-Stop, Network Failure, Timing Failure, and Byzantine Failures; each of which I will discuss separately. The first of the four failures in the distributed system is Fail-Stop and this is when a halting failure occurs with a type of notification to other components, and this can be when a network file server is in the process of telling its clients it is about to stop executing, and in the process the internal state and the contents connected to the volatile storage can be lost. The second type of failure in a distributed system is network failure, and this can keep processors from being able to communicate with one another. One of Two problems that come up are one way link and which can lead to problems such as the processors slowing down, this can cause one processor not being able to receive messages from the other processor. The second problem that arises is Network partition and occurs when the connecting line of two sections of the network fail, and it can causes a group of two processors to be able to communicate with one another but not with another group of two processors; this can lead to the two groups of processed downloading a file in different ways leading to the file inconsistent among all processors. The Third type of failure in a distributed system is Timing Failure, is the process or part of one that fails to meet its limit set for executing the process, message, clock drift rate, and clock skew on time. The timing failure causes components to respond with the correct value that is outside the specific interval meaning that it is too soon, or too late. Also overloaded processors can be hit with excessive delays even if the correct values are produced, and most timing failures are only in systems which have timing constraints and computations. The Fourth type of failure in distributed systems is Byzantine Failures occur when or during an execution of an algorithm, and when this failure occurs it can cause the system to respond in an unpredictable way by processing a request in the incorrect way, and corrupting local state in addition to sending an inconsistent response to request, and in a way failing to even receive the request. This can occur when an output of one function happens to be the input of another it causes small round off errors in the first function that could then lead to larger errors in the second function. Out of the four failures that can occur in a distributed system two of them can also occur or be in a centralized system, and the failures are Fail-Stop and Network Failure. I would assume that these two failures will fault the same way in a centralized system as they do in the distributed system. Of the two out of four failures that can occur in a distributed system the first of which to isolate and fix is network failure, this can be done by spoofing the network. Using the spoofing technique a probe can be sent down paths to find if the failures are on a forward path or reverse path. The way the network failures can be addressed and fixed are by using a network failure detection and recovery in a two-node by using windows server 2000 cluster that runs a sophisticated algorithm which can detect available network interfaces along with the plug and play function to detect disconnected cables used to connect the network, as well as connectivity issues between the network adapter and the hub or switch. Using windows server 2000 cluster can help detect network failures and lead to resolving them. The second out of the four failures in distributed system is t he Byzantine Failure and the way to isolate and fix the failure is by using a solution called Practical Byzantine Fault Tolerance which is an algorithm that can provide high performance Byzantine state machine replication, allowing for processing thousands of request a second with the use of sub-millisecond increase in latency. Another way to help isolate and fix Byzantine Failures is using a redundant system that can use or migrate of mask the effect of a limited amount of faults through redundancy, this can lead to the detection of faulty nodes and identifying and isolating them before they can harm from failure. When it comes to dealing with a distributed system, an organization can run into failures, four of which were described in this paper and they are; Fail-Stop, Network Failure, Timing Failure, and Byzantine Failure but this also includes others that were not discussed. Never the less, these failures can and will occur, it is up to the organization of company running their system to identity these failure risk, isolate them if they are found, and of course they action to guard against and fix issues of failures so it will not cause irreversible damage and harm that can lead to loss of information and time. These failures are inevitable but knowing about them and how to use fault tolerant protocols will indeed safeguard a distributed system Reference http://www.combofix.org/what-is-network-spoofing-and-what-are-the-types-of-it.php http://support.microsoft.com/kb/242600 https://www.usenix.org/legacy/event/hotdep06/tech/prelim_papers/haeberlen/haeberlen_html/