KIN 4310 1st Edition Lecture 23 Outline of Last Lecture I Dietary Assessment Methods II 24 Hour Recall III Food Record Diary IV Unweighed Estimated Food Record V Estimating Food Portion Sizes VI Food Frequency Questionnaires VII Diet History VIII Duplicate Food Collection IX Summary X Definitions XI Measuring Physical Activity XII Intensity XIII Physical Activity Level PAL XIV Methods to Assess PA XV Accelerometers XVI Heart Rate Monitors XVII Direct Observation XVIII Pedometers XIX Physical Activity Logs PALS These notes represent a detailed interpretation of the professor s lecture GradeBuddy is best used as a supplement to your own notes not as a substitute XX Physical Activity Questionnaires XXI Summary Outline of Current Lecture I Review Questions Current Lecture I Review Questions a What is a hypothesis i A hypothesis is 1 A widely understood fact of nature is a law 2 An explanation of some observed phenomenon that is supported by large amounts of convincing evidence is a theory 3 An educated guess b What is statistics i Statistics with an s is not pleural It is a set of tools used for planning experiments collecting data summarizing data and drawing inferences from data it is the whole philosophy behind it We learned specific statistical tests t test one sample z test anova etc In addition to those there are at least 40 or 50 more c What is a population A census A sample i Population is context specific It is the collection of all elements to which hypothesis applies ii Census an attempt to collect and measure every individual in a population iii Sample a sub set of the population that you select to be a part of your study d Descriptive Statistics i The mean is 9 18 ii The standard deviation is 1 54 iii It is bimodal 1 Modes 10 4 and 12 iv It is positively skewed v Note A descriptive statistic is a number that describes the sample like mean sd mode shape of distribution skewed symmetrical etc e Inferential Statistics i Note Inferential statistics is what is used to test hypotheses You have a population from which you draw a sample and hopefully it is diverse like the population and is representative then you will do experiment procedure or make your observations measurements then you will calculate probabilities to determine critical values p values etc The probabilities allows us to makes these observations f Describe 3 different types of sampling i Know all 5 g h i j k l m n o p q r s t ii Systematic random cluster convenience stratified Which type of sampling is best i Random sampling Why i It eliminates bias and it generally represents the population but it is not the easiest thing to do because you need a master list and sometimes it isn t even possible to administrate A group of UH researchers went to a local meal center and invited every fifth meal recipient to participate in a survey This is an example of a i D systematic sampling Name 3 measures of central tendency i Mean median mode Name 3 measures of variability i S d range variance What is a frequency distribution How is it usually illustrated i You are given a set of numbers you put it in order determine how many belong to a certain class bin It is a table of numbers The left column is the range and the x is frequency There is no overlap in the bins and every piece of data fits into a bin Know the tails of positive and negatively skewed graphs Which of the following statements is true of data with a nominal level of measurement i You can calculate a mean nominal doesn t use numbers ii You cannot generate a frequency distribution you can generate a frequency distribution but not a histogram because the x axis is not numerical iii Paired with another variable you can use it to calculate a linear correlation coefficient you need interval or ratio for r iv You can calculate a mode v Note a nominal level of measurement means in name only It is not ordinal so you cant put measurements in order With a histogram the x axis has to be numerical What is an observational study i No independent variable You just go to measure things You don t want to influence what you re measuring at all Good for helping to describe samples What is an experimental study i 2 characteristics the variable and you can draw a causal inference A causes B can only be done with an experiment not an observation What is an independent variable A dependent variable i Independent variable variable in experiment that investigators control ii Dependent variable all things that you measure the output What is a longitudinal study i A study where you look at the changes over time Like a weight loss study What is a correlation i An association with the general population It is not a causation The variables are related to each other What is represented by the linear correlation coefficient r i It represents strength and direction of correlation Interpret the following u v w x y z aa bb cc dd ee ff i rxy 0 71 1 Pretty moderate to strong correlation between x and y ii rxy2 0 50 1 r2 is the correlation of determination A number from 0 and 1 that is explained by the relationship to the other variable This means that 50 of the variance in x is explained by y iii rxy2 1 1 All of the variance in x is explained y its relationship with y It s the perfect correlation But you don t know if it is positive or negative Just that if you know x you can predict y if you know the correlation between the two The scatter plot on the previous slide showed i B a weak positive correlation What is reliability i Reliability and validity are properties that describe different measuring tools or instruments Reliability is the observation made using instrument will be close to the true score Anything between observation score and true score is error Define 3 kinds of reliability i There is 4 test retest interrater parallel forms internal consistency What is validity i Has to do with how you are applying them like the context Meaningful useful and appropriate Define 3 kinds of validity i Content criterion and construct The Physical Fitness Test recognizes students for their level of physical fitness in 5 events curl ups shuttle run endurance run push ups and V sit Is a child who scores well in one event likely to score well in the other four events This is a question of i B internal consistency reliability Match up i Test retest reliability every time I do a 1 5 mile run test my performance varies a lot ii Interrater reliability the number of calories determined by a 24 hour food