KIN 4310 1st Edition Exam 4 Study Guide Lectures 23 24 Lecture 23 April 28 A hypothesis is an educated guess A theory is an explanation of some observed phenomenon that is supported by large amounts of convincing evidence A law is a widely understood fact of nature Statistics is a set of tools used for planning experiments collecting data summarizing data drawing inferences from data and the whole philosophy behind it A population is context specific and is a collection of all the elements to which the hypothesis applies A census is an attempt to collect and measure every individual in a population A sample is a sub set of the population that you select to be a part of your study There are 5 different types of sampling and they are systematic random cluster convenience and stratified Out of the 5 random sampling is the best because it eliminates bias and generally represents the population There are 3 measures of central tendency and they are mean median and mode There are 3 measures of variability and they are standard deviation range and variance A frequency distribution is a table of numbers and is illustrated usually by histograms An observational study just measures things and does not attempt to influence whatever is being measured An experimental study has two characteristics one being that you can draw a causal inference from the data and you have a variable that you can manipulate An independent variable is the variable that the investigators control and the dependent variable is what the investigators measure A longitudinal study is a study where you look at the changes over time A correlation is not causation It is an association with the general population The linear correlation coefficient r represents the strength and direction of the correlation Reliability is the observation made using the instrument will be close to the true score There are four kinds of reliability and they are interrater parallel forms test retest and internal consistency Validity has to do with how meaningful useful and appropriate something it There are three kinds of validity and they are content criterion and construct A null hypothesis is the negation of a claim and is what we always assume is true at the beginning of a study because it is best to be skeptical until we have strong evidence The research hypothesis is the claim and is the formal statement of the claim A test statistic is the number that you get after crunching down all of your data and it is used to make a decision on whether or not to reject the null A critical value is the partition between the red and blue zones and is determined by the tables in the back of the book The level of significance is the probability of a rare event known as alpha Lecture 24 April 30 A type I error is a false positive and is alpha while a type II error is a false negative and is beta A p value is the probability of getting your test statistic or one more extreme that it if the null hypothesis is true A one sample z test is a hypothesis test that compares a group or a sample to a population You need the sample size population mean and population standard deviation to perform a one sample z test You use Tabl B1 to determine the critical values of z A t test is a general test to determine if there is a significant difference between two groups The t distribution is symmetrical about zero and goes from negative infinity to positive infinity and it changes shape according to degrees of freedom Degrees of freedom represents the size of your data An ANOVA is an analysis of variance and it is a hypothesis test that determines if there are differences between two or more groups using the test statistic F The distribution is asymmetrical and is positively skewed We find the critical values of F by looking at table B3 Variance between samples concerns how much the group means vary with respect to one another Variance within samples takes each individual sample and concerns how much the individual sample varies with respect to its own group mean We test hypotheses concerning the linear correlation coefficient by looking at the paired data making a scatter plot calculating r and then looking up the critical values using Table B4 Aerobic Fitness is the ability of the body to take in O2 from the lungs transfer it to the circulatory system and use it in the muscles to produce mechanical energy We measure it in a bunch of different ways with VO2max being the most accurate The biggest pro of a VO2max test is that it is the most reliable although it is expensive and can be risky There are two types of fat that we discussed essential and storage We can measure body composition through direct techniques such as the DXA scan MRI and CT The DXA is the gold standard because it is fairly cheap and very accurate and low radiation Some indirect techniques are skin calipers hydrostatic weighing bod pod bioelectrical impedance and air displacement An advantage of skin calipers is that it is low cost and is pretty accurate The four methods to assess diet are food diary food frequency questionnaire duplicate food collection and a detailed diet history The method that is prospective is the food diary and a duplicate food collection The cheapest method is the questionnaire and the most reliable method is the duplicate food collection Physical Activity is any movement that burns energy Exercise is a sub set of physical activity that is planned purposeful and repetitive We can measure physical activity objectively with a heart rate monitor accelerometer or a pedometer because they take human interpretation out of it There are a bunch of things that can be used for a self reported physical activity score one being the SRPAQ Prevalence and incidence deal with morbidity Prevalence is the overall amount of people who have the disease at one point in time while incidence deals with the new cases Variables like morbidity mortality life expectancy at birth disease rates and obesity are used to represent the health status of a group nation Disease Adjusted Life Years are the number of years that you would prematurely die the years spent in a diseased state This number is used in cost benefit studies