|· Portal||Help Search Members Calendar|
|Welcome Guest ( Log In | Register )||Resend Validation Email|
|Welcome to Auxillary Literature. We hope you enjoy your visit.|
You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.
Join our community!
If you're already a member please log in to your account to access all of our features:
Posted: Jun 6 2009, 05:18 AM
Member No.: 2
Joined: 2-May 09
We'll look at common measures of disease frequency for public health purposes, and we'll notice that there's a crude death rate and there's a live birth rate. The crude death, or mortality rate, equals the total number of deaths from all causes per 100,000 people. Look at the size of the denominator. Now, the denominator is the same for the cause specific, age specific, and race specific death rates. For cause specific death rates, the numerator would be the total number of deaths from stroke or asthma per 100,000 people. Age specific crude death rate would be the total number of people age 20-29 years old from all causes per 100,000. And the race specific would be total number of deaths from all causes among african americans per 100,000.
Now, let's look at the live birth rate. The numerator is the total number of live births and the denominator is 1000 people or sometimes a total number of live births per 1000 women of child bearing age. Now let's look at infant mortality rate. Here, the numerator is the total number of infant deaths over the denominator – total number of live births. Here, we're defining infancy from birth to one year of age. Let's look at some other race. The attack rate consists of the number of cases of disease that develop during a defined period over a denominator of the number of population at risk at the start of the period. So, if we're looking at the attack rate of the h1n1 virus in the city of st. louis, we'd be looking at the total number of cases of h1n1 confirmed disease in 2009 divided by total number of population at risk during period which would be the total number of people in st. louis at the beginning of 2009. Note that the attack rate is usually used in context of infection of disease outbreaks.
Next, we'll look at the case fatality rate, which is number of deaths for a defined period of time over a number of cases of diseases. So, if there are 100 disease cases and there are 5 deaths, the fatality rate would be 5/100, or 5%. The survival rate represents the number of living cases over a defined period of time over the number of cases of disease. The operative term in the numerator is living, and the denominator is represented by the total number of cases.
Now, here's some questions. The Boston Globe report on Celebrity skiing deaths. Although skiing has inherent risk, it is not more dangerous than other common activities according to the Boston Globe, the nationwide comparisons below do not reflect different numbers of participates. The Globe warned the reader that the comparisons do not reflect the differing number of participants, what does that mean? Means that the denominators are not comparable at all. So see the next slide for the stats and then discuss online with each other – is the above statement accurate? Does skiing have inherent risks, but not more dangerous than any other activities. And what are the numbers in the following charts? Can you determine, for example, the cumulative incidence of drowning in swimming, boating, or water sports. Can you determine the prevalence in any one season? Now, in reference to these accidental deaths, how does skiing compare to other sports? When discussing, be sure to discuss the importance of the missing denominators.
Now, after we've looked at the number of deaths in various seasons, etc. with inadequate denominators, we'll move onto the next discussion exercise, which is to practice measures of disease frequency.
What you'll be doing is state which type of mathematical parameter – ratio, proportion, or rate, and which type of measure of disease frequency – such as cumulative incidence, incidence rate, prevalence best describe each of the following situations on the next couple of slides. So, you're going to look at the percentage of infants enrolled in daycare center who contracted impetigo during course of an epidemic. That sort of gives away which parameter you'll be looking at, but it doesn't cover the measure of disease frequency. The next one also gives away the mathematical parameter, and the next one asks for the number of colds experienced in the year per 1000 people. Tell me what parameter, what measure of disease frequency. The next one asks for a percent, so the parameter is provided, of the deceased males found to have had prostate cancer in their autopsy. But tell me the disease frequency that is. And then there are a number on the next slide, number of still births per 1000 life births, percent of people with malaria that are resistant to treatment with chloroquine, the number of newly diagnosed brain tumors in a year per 100,000 children, the percent of live born infants with cardiac malformation per 100,000 live births. This exercise is conceptual.
But in part two, you're going to apply epidemiology thinking. This is an investigation that began on jan 1st, 1999, in which from a population of 1000, four individuals were found to have the disease under study. During the subsequent year, jan 1st onward, six additional new cases were found. There were a total of ten cases, and there were six deaths during the year. For the ten cases, the following diagram indicates the time of case recognition, period of observation, the vital status of the patient at the time of study. And an arrow indicates at the start of the diagram that patients 1,2,3, and 4, actually had the disease on jan 1st, the date the study began. Here we have the stats, and you'll see that patients 1-4 had already been diagnosed sometime prior to jan 1st, 1999. and that case number 5 diagnosed in jan, as well as case number 6, who died, but case 5 was alive at end of year. Case 7 was alive, 8 died. 9 was alive, and 10 died around november. Assume that the remaining 990 members in the study did not become ill or die during the year of observation. Next slide – answer the questions. What is prevalence on jan 1st, july 1st, and december 31st? Also, what is the cumulative incidence of the disease during 1999? And then on the next slide and it's asking for the population mortality rate and note that it is the cumulative incidence, or rather it is the rate that you can derive from the cumulative incidence. And then determine the case fatality rate during 1999, and which of the above measures would be the best indicator for which of the following purposes? Determining the effectiveness of new treatment? Evaluating the effectiveness of new treatment? Or Estimating the needs for medical facilities in order to treat the disease? Don't forget that the measures you are looking at are prevalence, incidence rate, and cumulative incidence, and which of these would be the best measures?
Now, just off the top of my head, I would think that such measures of disease frequency could also be applied to various projects, such as proportions of patients at go on the insulin pump, proportion of patients at go using online induction formulae, or any number of different clinical projects. So while we're talking about disease frequency, we're talking about disease broadly. Disease can be risk for hypoglycemia, risk for bleeding, obesity, etc.
There are two main options for comparing data. We can calculate the ratio for two measures of disease frequency, or we can calculate the difference between the two measures. The first step, however, is to set up a contingency table. Most often, it's a 2x2 table. Now, we can use this type of table for cumulative incidence and prevalence data. In the table we're looking at whether the disease was present, and responses are either yes or no. And whether the exposure was present – responses also yes or no.
Now, when we use person/time in the denominator for incidence rates, then we can also use this 2x2 table to evaluate the presence or absence of disease per person years or per pack years of cigarette smoking and then we can compare it to exposure – yes or no.
When we look at the rate of disease to the risk of exposure, then we're looking at a rate to risk ratio, and we can also call this a relative risk. What is the relative risk of a disease in a population given exposure to factor x? On the next slide we see the formula for relative risk and cumulative incidence and for incidence rate. So, let's look at the relative risk formula first. It's the rate or risk in the exposed group divided by the rate or risk in the unexposed group. Let's look at the cumulative incidence ratio and that's the cumulative incidence in the exposed group over the cumulative incidence in the unexposed group. In the incidence rate, we have the incidence rate in the exposed group divided by the incidence rate in the unexposed group. What do those A's and C's refer to?
You'll the see the A cell, C cell, and A+C cell. You'll also see the B cell, D cell, and B+D cell. So, when you see A+C, you know that the numbers in A are added to the numbers in C. Same with B and D.
On the next slide, we approach the question of what is the purpose of the relative risk? Well, relative risk gives information on the relative effect of the exposure on the disease. It tells you how many times, higher or lower, the disease risk is among the exposed as compared to the unexposed population. And relative risk is commonly used in etiologic research. If anyone is studying the etiology of a nursing diagnosis, then you'll see that relative risk is one of the formulas you'll be using.
On the next slide we see that if the relative risk is 1, there is no association between exposure and disease. The disease is just as common in the exposed and the unexposed populations. However, if the risk is 2, it's 2 times the risk in the exposed population compared to the unexposed population. Now, what's interesting about a risk of 2, is that if you have a small sample size, a risk of 2 probably indicates that there is probably confounding going on. The larger the sample size the more impressive the risk of 2 is. But a 2 in a smaller sample size may indicate confounding.
Let's say the relative risk is 1.6, then that means there is 1.6 times the risk in the exposed group compared to the unexposed group. This is can also be phrased in another, in that there is a 60% increase in the risk of disease in the exposed group. How did they determine this number? Well, they subtracted from the relative risk of 1.6 the relative risk that shows no association – 1. So, 1.6 – 1 = .6. When .6 is converted to a percent, it is 60%, so 60% increase. Now that 60% sounds pretty impressive, but if you're dealing with a rare event, then a 60% increase is not all that great. For example, if you have an event that occurs in 100000 cases, then a 1.6 increase is not going to be a tremendously large increase. Now, you will have 1.6 events in 100,00 cases. You can say there is going to a 60% increase, but that sounds somewhat misleading, because it sounds like there has been a huge increase when there actually hasn't. On the other hand, it's easier for laid people to understand that there is a 60% increase in the risk of the event than it is for laid people to understand a risk of 1.6. Now, we'll go ahead and apply these formulas to a cool hard study of hypertension and cardiovascular morbidity and mortality. And a cool hard study is a study in which a population of people are followed over time. For example, in the nurses' house study, a large number of nurses were followed over a period of 20-25 perhaps up to even 30 years now, and the study was conducted by researchers at Harvard.
Anyway, in this slide, we're looking at the exposure to hypertension, the risk factor, of a non-fatal heart attack – the disease.
Now, we're going to look at the relative risk, and we're going to be calculating the relative risk associated with the cumulative incidence rate in the exposed and unexposed population. So, this is a relative risk of cumulative incidence rate. So, relative risk = the cumulative incidence in the exposed over the cumulative incidence in the unexposed population. So, if you go to the previous slide, you'll see that in cell a, there were 117 people who developed a heart attack in the exposed population, but there were 125 people who developed a nonfatal heart attack who were unexposed. However, the denominators are considerably different in size. The 117 had a denominator of 13422, that is the total number of people with hypertension in the population. Now, in that population of nurses, there were four more people who did not have hypertension. In fact, 106541 nurses did not have hypertension. So, once the relative risk of hypertension being associated with nonfatal heart attack among a population nurses is .00872 divided by .00117 which equals a relative risk of 7.5, the relative risk of hypertension leading to nonfatal heart attack is 7.5 times greater than the nonfatal heart attack occurring in the population that does not have hypertension. Another way of saying that is that women with hypertension have 7.5 times the risk of having a nonfatal heart attack as the people who do not have hypertension.
On the next slide, we have a 3x2 contingency table. We have 3 rows, indicating the severity of magnetic field exposure and we have leukemia yes or no. Then we have a column to the right of the total column marked relative risk. Now, we're all exposed to magnetic fields daily, but most of us are exposed to low magnetic field levels. So, the low group is actually the control group. Now, look at the relative risk of leukemia given magnetic field exposure in low, medium and high levels. In the low group, it's 1.0, in the medium, it's 1.23, in the high, it's 1.33. So, the interpretation is that children exposed to medium levels have a 23% increase risk in leukemia, etc as compared to children exposed to low magnetic field levels.
On the next slide we see that there is another of way of comparing the disease occurrence among exposed and unexposed populations, and that's by looking at differences in population, by subtracting one rate from another rate.
On the next slide, we show what these formulas are. We're looking at the risk rate difference, and it's also called the attributable risk rate. Now, the risk difference or rate or risk in the exposed population minus the risk rate in the unexposed is the overall definition, but how do we find it? Well, if we're using cumulative incidences, then we subtract the incidence in the unexposed population from the incidence in the exposed population. Now, the formula for cumulative incidence in the exposed group is A/A+B and then the incidence formula for the unexposed is C/C+D, and since you're subtracting the cumulative incidence in the unexposed population from the incidence in the exposed, then you're subtracting C/C+D from A/A+B. For incidence rate, you're looking at the incidence rate in the unexposed population subtracted from the rate in the exposed population, and that formula is A/Person time in the exposed minus C/person time in the unexposed. Now, unlike the relative risk, where 1.0 shows no association, if you're subtracting differences, then there's no association if the answer is 0.
On the next slide, what's the purpose of the risk rate difference? It gives information on the absolute effect of exposure on disease occurrence. It also gives information on the access disease risk in the exposed group compared to the unexposed. It finally gives information on the public health impact of an exposure. That is how much disease would be prevented if the exposure itself were removed. This assumes that the exposure is of course related to the disease.
On the next slide we look at the nursing study again, and let's look at the exposure to hypertension and its effects on nonfatal MI. If we're looking at the relative difference, then we're looking at the cumulative incidence in the unexposed population subtracted from the cumulative incidence in the exposed. So, we're looking at the cumulative incidence in the unexposed being 125/106541, and we're subtracting that number from the cumulative incidence in the exposed population, and that's 117/13422, and the answer is .00872 - .00117, and that is 755 persons per 100,000 is the risk attributable to hypertension in this nurse's health study. Now, another way of interpreting that is to say that the excess occurrence of nonfatal heart attack among these hypertensive women is 755 per 100,000, or if hypertension causes nonfatal heart attacks, then 755 cases of nonfatal heart attacks per 100,000 women could be eliminated if hypertension were treated.
Now, let's go onto the next slide, in which we're comparing relative risk and relative difference. We're looking at the annual mortality rate per 100,000 associated with lung cancer given the smoking status with the individual. Now, if you look at the last and second to last rows, you'll notice that the last row refers to the difference, and the third row refers to the relative risk. The relative risk of developing lung cancer, given that you're a smoker, is 14. So, the risk is 14 times great among smokers compared to nonsmokers. Now, the chance of acquiring coronary artery disease is only 1.6 times greater among smokers than nonsmokers, but let's look at the relative difference. If we eliminated cigarette smoking, we'd save 130 lives per 100,000 per year. On the other hand, if we eliminated smoking, we'd be saving more than twice that number in terms of coronary artery disease deaths. So, in conclusion is smoking is a much stronger risk factor for lung cancer but assuming that smoking is related to both diseases, the elimination of cigarettes would prevent far more deaths from coronary artery disease. Can you think of any reasons why? One clue is disease frequencies so you may want to go off in that direction when seeking an answer.
Now, on the next slide, we're looking at population risk or rate differences, and it's abbreviated PRD. Now the purpose for calculating the PRD is because it measures excess disease occurrence among the total population associated with exposure. It helps evaluate which exposures are most relevant to the health of a target population. Now, what is that target population? It depends on what you're studying. If you're looking at the nurse's health study, then the target population is the total population of nurses enrolled in the nurse's health study. With magnetic field exposure, then you're looking at the total number of people in the study exposed to low level, medium, or high level exposure.
Now, just because we're adding another letter to the formula doesn't mean the concept is any more difficult. Now, we're looking at the population risk or rate difference. Two formulas are available for calculating the PRD. The first is where PRD = relative difference times population who is exposed. The second is to subtract the risk or rate in the unexposed population from the risk or rate in the total population.
Next slide, we see that when you multiply the rate difference times population exposed in the first PRD formula, the answer is .00085, and we'll interpret that in a couple minutes. In the second formula, the answer is 85 per 100,000, but what is the interpretation? Well, the interpretation is that hypertension results in an excess incidence of 85 per 100,000 nonfatal MI's in the total population, or if hypertension were eliminated, there would be 85 cases of nonfatal MI's that could be eliminated among the total study population. Now, it's important to realize that the population risk or rate difference is dependent upon the prevalence of the exposure in the population. It's important to realize too that a weak risk factor in terms of its relative risk, let's say it has a risk of 1.6 or 2.0, that's quite prevalent in the population, that could account for disease incidence than a stronger risk factor that is rarely present.
Now, let's look at calculating measures of comparison for cigarette smoking and lung cancer in the population, which is based on Dall and Hale's study from 1964, at which time the prevalence of smoking was 56%. The death rate from lung cancer in smokers was .96 per 1000 per year, so almost 1 per 1000 per year, where the death rate from lung cancer in non smokers was practically zero.
On the next slide we see that the rate ratio was 13.7, meaning that the rate of lung cancer in those who smoked was 13.7 times greater than the rate of lung cancer in ones who did not smoke. There was an 89% difference in the rates of lung cancer attributable to smoking than to nonsmoking. In that number, I'm looking at the rate difference, and the rate population difference showed that if smoking were completely eliminated that for every 1000 people, one half of a person would be saved.
Now, here's an exercise to practice. In 1976, the American legion held a major convention in Philadelphia. And this became the sight for one of the major disease mysteries solved by epidemiology. It seems that many people, legionnaires mainly, in the hotel, became ill with a pneumonia to which many succumbed. And since it was summer time as opposed to the regular pneumonia season, and since the death attributable to the pneumonia were so much higher than usual, major studies were initiated to find out the etiology of the disease and its recommended treatment. Here is some of the data from Philadelphia. The convention status of those who developed and those who did not develop legionnaire disease were evaluated. And as you can see from this slide, the delegates developed legionnaire disease at a much higher rate than non delegates. So, compute the cumulative incidence of legionnaire disease among the delegates and non delegates and what kind of measure. Next, develop the cumulative incidence ratio of the disease among delegates compared to non delegates, and state the meaning of this ratio. Next, count the cumulative incidence difference of the disease among delegates and non delegates.
Then, on the next slide, continue to analyze the data that was collected on the legionnaires during and after the convention. The epidemiologists wanted to know if legionnaires only developed an illness who remained at the hotel not only at the convention but their place of residence during their stay in the city. Or if legionnaires also occurred among those who stayed in other hotels and came to the convention and then returned to their own hotels.
With this data, calculate the cumulative incidence of legionnaires disease among residents in hotel a and residents in the other hotels. Also, calculate the relative risk associated with hotel a and the risk associated with the other hotel, and then calculate the relative difference between hotel a and other hotels.
Okay, welcome to using epi-info software for epidemiological studies and investigations. If we move on the next slide, let's look at the objectives. One – to become familiar with epi-info software. Two – become aware of the data available for a cholera outbreak analysis, and this is the example that is available on the epi-info cdc website, and we're going to use that example to learn the software. One thing to note is that this is just to get familiar with the technology. This week, I want you to just download the software and surf through the various pages. See what's there, become familiar with what's there, and I just want to stress that you need to get familiar with the software and you will not get familiar by just opening once and acquiring the data. It's going to take multiple times to download the software. The software is pretty easy; however, you need to be aware that if you have a different versions (ie – vista or xp), you may have to download the software with extra steps. If so, contact the epi-info section on the cdc website.
Onto the next slide, what is epi-info? Epi-info is a statistical software package for epidemiological studies, which you may use in your own clinic or whatever type of career you have chosen. It is free, and you can locate it on the cdc website, or just google “epi-info.” One thing about it – enjoy, because the taxes are paid for, you don't have to pay for it.
Many stat softwares cost a lot of money. There's different softwares that I've purchased that have cost over 1000 dollars, but if you're in an academic center you can get it for 500 dollars. There are other softwares available, but you have to be careful when using these softwares because some had limitations. One of the biggest limitations is that it doesn't offer all of the statistical analysis that you may require, and I have found that if it doesn't have that certain statistical analysis, professors would have used something else.
Another scary encounter that I came across is I ended up with wrong answers on a software off the Internet. I repeated it on another software and also did it manually to figure out the answers are wrong. Once again, the more you use the software, the more comfortable you will be with the software.
Moving on, epi-info software download, so what I've provided on this slide is how to download the software, and I've typed out step by step info on what you need to do. So, the first part is the website. Then, in the bold and underlined, I give you the command, click on the download at the top of the page. Then, click on full download. Click on install, recommended, and that's located near web install, and then simply click run. You might encounter a statement that the publisher does not recognize the program. This is the message that I encountered on my laptop, but I did continue at this point, and I've had no problems that I'm aware of. Click next three times, then, and keep clicking until you reach finish. Then, your software is loaded. And you should see an icon on your desktop. Now, part of the download will be the training modules, and there are 12 epi-info modules. We're only going to be working with 5 though, and those modules will answer the questions and help you investigate the outbreak investigation.
Next slide, let's talk about the modules. The first module will provide a description of the five out of the twelve modules. “Welcome to the using epi-info in an outbreak investigation, provided by the cdc in atlanta, georgia. The purpose is to introduce you to the features and functionality of the epi-info software as it pertains to outbreak situations.” There is a sixth module that I'm adding, so we're going to look at the five modules from the training site within the cdc web site, but I am adding a sixth module for sample size estimation and poiwer analysis and i will explain that to you in a later presentation. But the reason you have to do an estimation is because you want to be sure that when you design a study, you want to make sure that you have enough participants to provide generalizable and reliable results.
Module one will be done for the first two weeks of the course. Spend about 30 minutes a day getting familiar with the software. Get familiar with the screens and terminology.
Next slide – epi-info intro continued. After completion of module two, so now we're on module two, introduction to cholera, the student will have reviewed the case study. This module provides a description of the problem, and they suspect a cholera outbreak. The information from module two is something you must consider when studying an outbreak. Do not underestimate the value of any piece of info that is described in module two. When you read this module, you need to read it closely, and identify important elements or characteristics that might lead to the answer in this investigation.
Next slide – we're now on module six. Before you analyze your data, you may need to perform some basic data management skills. What's data management? It takes a little bit more time than you might think. Data management is when you want to have your data in your spreadsheet in a form in which it can be statistically analyzed. Once it's in this format, you must be aware of the fact that you might have to create new variables. For example, if you have an age variable, you might want to change it from a numerical variable to a categorical variable, so you may want to look at different age groups. For example, anyone less than 17 is group one, 17-25 is group 2, 25-35 is group 3, etc., because you may find differences in who contracted cholera by age group.
Next slide – and this is module 7. Module 7 is describing the epidemic by time, place, and person. Here you'll be performing the actual statistical analysis. There are twelve questions, 2 you will answer in module six, basic questions about frequencies about people involved in epidemic, while in module 7 you're going to perform a few more analyses. You know that you are going to have the answers, and they are provided because the purpose of this presentation is to make you familiar with the software, so what I want you to do is submit the output, so when you get a question and it asks what the frequencies are, I want to see your output. If you got the wrong answer, maybe you clicked on something incorrect, so go back and check if you get a question wrong. There might be something wrong with your output.
Next slide - the last module that you will be viewing is the conclusion of the cholera outbreak investigation. So, the crucial results of studying the outbreak should be concluding how the disease may have been spread, and this may involve several factors. Using this information to implement additional control in prevention measures, communicating the findings to state holders, and making recommendations to control these factors. Read the conclusions. Any investigations become a useless exercise unless relevant conclusions are drawn as well as acted upon. You will be expected to list the work you do so what I'm asking here is that when you read module 11, I want you for the last assignment to send me a list of what you're going to do with the information from your analysis.
Going on to slide 11, the overall objectives of the study. The overall purpose is for you to become familiar with epi-info software when investigating an outbreak. As far as grading, this will be on a pass-fail basis. Since you have all the answers, it'll be pass-fail because I just want to see that you did actually do the exercise.
The last presentation is the sample size estimation which I mentioned previously. This will be to demonstrates how to determine adequate number of participants to provide legitimate and reliable results.
Okay, access to the modules. I'm not going to read each item here, but these are the steps that you choose so that you can access the modules, and this meant to be a step by step, calm approach. You can see that I have provided the site, and you can choose tutorials to help you out. You will then choose the epi-info cholera outbreak tutorial. And this uses a case study based on an actual investigation of an outbreak in Uganda. And the epi-info software is used as a tool to perform data entry and analysis activities commonly encountered in an outbreak investigation. And you see in red, more, and you click on that to receive more information on the investigation. Once you do that, you will choose 'cholera outbreak' using epi-info for windows in an outbreak investigation. This introductory training for the software can be used as a self study for individuals for strong basic computer skills or as a classroom based training tool. Provides 18 hours of instruction or 3 days of classroom training. You probably won't require that much time because you won't be going through all twelve modules. The target audience includes epidemiologists who help manage and analyze data.
Slide 15 – the following materials are available for download. You should download the data that's under database, and then you would download the training session modules, so then you'd be doing 4 downloads. The software, and then the three things listed on slide 15.
Obtaining data for the cholera investigation. Module 1 provides step by step directions. Do not be discouraged if the steps do not lead you to the data. There may be compatibility issues depending on which version of windows you are using. If that's the case, contact one of your colleagues or email me.
The next two slides are slides that just take you step by step on how to obtain the database and the training folder. It will be automatically labeled a training folder.
You should now be able to view win zip on your computer screen. The modules, once again, that you will be using for this training session are modules 1, 2, 6, 7, 11, and 12. 12 is optional. You may not have time for it, but in the future, you might want to look at it because it's pretty interesting.
Slide 20. There are several appendices that epi-info provides. This is a list of the appendices you need when performing data management and analysis when using epi-info software. So uh, make sure you look at appendix A. That was the questionnaire given to the participants in this geographical area of people who are getting diagnosed with cholera. So you want to look at it because it's important. Appendix B has guidelines for specific diseases and conditions. There's a code sheet because all the data elements will be coded. Appendix F is where you will find the answers for the output assignment. Evaluation of assignments – 2 in module 6, 10 in module 7.
Next slide, for those of you who haven't performed any statistical analysis before, you might want to know what output is. It is the statistical computations you requested to address the various questions required to investigate the outbreak, and it can be in many forms – graph, table, map. It's dependent on what you choose. The output must be in a descriptive format. You will have the interpretation of the output and the answers in appendix F.
Next slide, once again, learning a new statistical software can be frustrating. Just keep playing with the software. This concludes the first presentation for week one. Thank you.
|Seattle "Red" Whitman||
Posted: Jun 7 2009, 08:06 PM
Member No.: 3
Joined: 2-May 09
Dont post any more of this fucking shit or i'll fucking kill your whole fucking family with a fucking blowtorch you god damn freak.
Posted: Jun 7 2009, 10:23 PM
Member No.: 13
Joined: 7-June 09
This shit is fucking stupid post this dumbshit again and im' going to fucking drive by your block and shoot you up you hear?
Posted: Jun 7 2009, 10:27 PM
Member No.: 2
Joined: 2-May 09
quit posting this stupid bullshit faggot or i'll ram a knive up your mouth you imbecile