Can some one help with this Lab Excercise.
Exercise 1: The Scientific Method
People learn about the world around them by observation. We see things, examine relationships, and discuss our observations with friends, family, and colleagues. These observations, without passing through the steps of scientific inquiry are anecdotal.
The scientific community goes beyond observing and speculating. The long-established logical sequence employed by scientists is the scientific method.
The first step is to establish something of interest which leads to a study. This culminates in the formation of a hypothesis. The hypothesis is a testable statement based on observations and relationships which have already been established. It is not based on speculation or guessing.
The hypothesis needs to be carefully worded. The intention is that it will most likely be tested again by independent researchers who will want to document the findings. It’s only when the results can be repeated that theories are developed to provide more widespread explanations.
Many scientists prefer to test a null hypothesis (H0), which suggests that the two groups are the same. If, in fact, the two groups are different, you can reject the H0. Otherwise, you have failed to reject the null hypothesis; this is perceived as support for it.
Testing the hypothesis involves the collection of quantitative data. Quantitative means that the information is numerical. This is necessary so that the data can be analyzed by an appropriately selected statistical test.
After this analysis, the investigators present results and draw conclusions. These conclusions include projection towards future studies necessary to expound upon and reveal more of the relationship in question.
For a study to be taken seriously, its results should be published in a scientific journal. Websites, magazines, newspapers, pamphlets, and blogs are loaded with reports that would never pass the scrutiny required to be published in a journal.
Studies reported in scientific journals have been thoroughly and critically reviewed by teams of unbiased experts in the field of study. Findings are generally not approved for publication unless they have been statistically treated and results can be supported by a minimum of 95% confidence.
There are two general types of studies, observational and experimental. Observational studies are very common in the field of nutrition. For example, there are countless longitudinal studies, taking place over long periods of time, during which thousands of people are being tracked but not experimented with.
One group might consist of people who consume a certain type of food, for example salmon. People in the other group may lack salmon in their diet. Data are collected for each group, for example rates of certain diseases.
The independent variable is the inclusion or exclusion of the food type. The dependent variable would be rates of a disease such as CAD (coronary artery disease). The hypothesis addresses the potential that development of CAD is dependent upon diet (the independent variable).
When interesting findings result from observational studies, researchers are encouraged to pursue experimental studies.
In this example, subjects in the experimental group would consume a certain amount of salmon (grams per kg of body weight) for a period of many years. Subjects in the control group would consume the same amount of an equivalent food, tilapia. Rates of CAD would be tracked over many years. Nutritional studies usually take a very long time to complete!
A sensible follow-up to a food study like the one in this example would be to refine the specific nutrient that might be responsible for a potential difference between the experimental and control subjects. What does salmon have that’s lacking in tilapia? Does a component of salmon offer protection from CAD or is there a trigger for the disease in tilapia that’s lacking in salmon? Is salmon “good” or tilapia “bad”?
As you can see, testing one hypothesis usually leads to the formation of more hypotheses and the need to conduct additional studies.
It is crucial to control variables other than the one being studied. All subjects participating in a study should have equivalent levels of general health, be as physically similar to each other as possible, be either non-smokers or smoke equally, and be taking no other drugs or herbal supplements.
Generally speaking, for any good study, the subjects should be as uniform as possible (age, gender, size….) to try to eliminate the potential for other influential variables. These subjects then need to be randomly assigned to the experimental and control groups; there can be no bias in assigning certain individuals to the experimental group.
Lab Report – The Scientific Method
Before proceeding, carefully read through this lab exercise. You should also read the section on the scientific method in your lecture textbook so that you have a solid background in the subject matter.
Based on observations that you have made in the past, or relationships that you might have heard about, you will develop and test your own hypothesis. Your challenge is to hypothesize a relationship between certain populations of people and some quantifiable (numerical) anatomical or physiological feature related to the course content.
Make sure that your hypothesis is testable and that you will be able to collect relevant data. Remember to control as many other variables as possible. All of your subjects should be in good health, the same gender and as close to the same age and height as possible.
Suggested Studies
The following examples are worded as null hypotheses, meaning that the statement predicts no difference between groups. If, in fact, there seems to be a difference, the null hypothesis will be rejected. Remember that the hypothesis is just a statement that you are testing; it is not what you think or predict.
You are free to test one of these suggestions or modify one to suit your own interests.
Smoking has no effect on resting pulse. Note that a lower resting heart rate is linked to good cardiovascular fitness and that the average rate for a healthy adult is 75 beats per minute (BPM).
Example: “People smoking a minimum of three cigarettes a day for a minimum of two years have similar resting pulse rates to non-smokers.” You would need to be careful to keep other variables, such as age and general health, constant.
You can compare body composition, reflected by body mass indices (BMIs) between two different groups of people. The contrast can be between athletes playing different positions, musicians playing different instruments, male vs. female actors; be creative!
For example:
BMIs are the same between professional golfers and soccer players.
BMIs can be determined from public record of heights and weights.
People on different types of diets, for example vegetarian vs. meat eaters, have similar BMIs. Once again, it would be very important to keep other variables, such as age and level of activity, as constant as possible when selecting subjects.
These are merely a few of countless comparisons that can be done. Your instructor will work with you to refine your hypothesis and help you set up your study. Unless you plan to test one of the suggested hypotheses, YOU ARE REQUIRED TO DISCUSS THIS STUDY WITH YOUR INSTRUCTOR BEFORE BEGINNING WORK ON IT. THERE WILL BE POINT DEDUCTIONS IF YOUR STUDY IS NOT SET UP WELL AS A RESULT OF NOT SEEKING GUIDANCE!
General Instructions
You will need a total of twenty subjects. Ten will be in each group.
All twenty individuals should be as similar to each other as possible regarding variables that might mask the one being studied.
For example, if you were interested in the relationship between diet and BMI you would not want to compare elderly vegetarians with thyroid disorders, who never exercise with young, active meat eaters.
You will collect the relevant data, display results in a table, and perform a statistical test to determine if you can reject the null hypothesis.
What to Submit – all submissions should be in the form of a Word document.
The following are required sections to be included in your report. Grading preference will be given to concise and well-structured sentences that do not stray off topic:
Introduction – This will include your hypothesis and how you decided upon it.
Methods – How you selected your subjects and gathered data.
Results – A description of your findings which includes a table displaying the data.
Analysis – A brief description of your statistical review, using the student’s t test described below. You should report the mean (average) and sample variance for each group, as well as the calculated t value. You may lose points if you don’t show your work.
Discussion – Does your t value allow you to accept or reject your null hypothesis? What are your thoughts regarding this?
Conclusions – Suggest a follow up study for further investigation. What would you have done differently this time?
Instructions for performing a student’s t test
This statistical test will allow you to assign confidence to a statement that you make regarding your null hypothesis. You will only be able to reject the null hypothesis, suggesting that your two groups are different, if the t value that you calculate is greater than 1.833.
In order to determine the t value you will need to calculate the mean (average) for each of your two groups. (The “x” with the bar over it is the symbol for mean.) Designate the group with the higher mean as group 1. The group with the lower mean will be group 2.
FORMULA FOR CALCULATING THE t VALUE:
Measures of Dispersion represent how widely spread observations are relative to the mean. It’s more difficult to demonstrate difference between two groups when there is great dispersion.
One measure of dispersion is variance. Variance is measure of how much variation exists in the population. You will need to calculate sample variance for each group to determine the t value.
This is done by subtracting the mean from each data value and then squaring that value as depicted by the equation below. These values are then added up and divided by n-1, which will be 9.
s2 = sample variance
Once you know the sample variance and mean for each group you can calculate the t value. Assign the group with the higher mean as Group 1 so your t value will be positive:
The numerator is the difference between the group means. The denominator is the square of the sum of the sample variances divided by n-1, which is 9.
As you can see, the t value will be larger with a bigger difference between the group averages and with less variance within each group, making the numerator larger than the denominator.
This means that even if the group averages are very different, if there’s also a lot of variation (scientists sometimes refer to this is “slop”) the t value will be small and the groups will not be significantly different.
THESE INSTRUCTIONS ARE ONLY RELEVANT IF YOU HAVE TWO GROUPS WITH TEN SUBJECTS IN EACH.
If your calculated t value is at least 1.833 you can reject the null hypothesis and you would be at least 95% certain that the two groups are different. If the t value is less than 1.833, you cannot reject the null hypothesis; the two groups are not significantly different.
8
