KIN 4310 Section 3 Review
F Statistic (2-way)
-F= -If there is a significant effect of treatment, then MSbetween will be large relative to MSwithin and F will be large -If F > Fcrit, then there is an effect of treatment, and we can reject the null hypothesis
Field Tests
-Field tests 1. 12-Minute Run/Walk Test -VO2max = (Distance in miles - 0.3138) / 0.0278 2. 1.5-Mile Run Test -VO2max = 3.5 + (483 / time in minutes) 3. 20-meter Shuttle Run (PACER) -VO2max = -32.678 + 6.592 (max speed in km/h)
Variance BETWEEN Groups (1-way)
-How spread out are the group means? -Total variance is based on all data from all groups -Variance BETWEEN groups represents the spread of the group means -We can calculate the sum of squares BETWEEN groups by subtraction: > SStotal = SSb + SSw > SSb = SStotal - SSw -Compute the Mean Sum of Squares: >MSb = SSb / dfb dfb = degrees of freedom between groups = number of groups - 1 = k - 1
Variance WITHIN Groups (1-way)
-How spread out is the data in each group relative to its group mean? -Represents the spread of each data point within its own group -Square the differences between every data point and its group mean, (X - X ) -Sum the squares, SSw = (X - X ) -Subtract each data value from its own group mean -Compute the MEAN SUM of SQUARES: MSw = SSw / dfw = (X - X ) / dfw dfw = degrees of freedom within groups = number of data points - number of groups = N - k
Assessing Body Composition
-INDIRECT TECHNIQUES: involve measurements outside or on the surface of the body to estimate what is inside -Hydrostatic weighing (dank tank) -Air displacement (bod pod) -Skinfold thickness -Bioelectrical impedance -BMI, WC, WHR -Girth measurements -BF%=(495/BD)-450
Health Status
-Individual Health Status >Observer (e.g., physician) >Self-report -Health of a population >Life expectancy at birth >Death rate >Prevalence of disease >Pollution
Air Displacement (BOD POD)
-Individual sits inside small chamber -Computerized pressure sensors determine the amount of air displaced by the person -Body volume is calculated by subtracting the air volume with the person inside the chamber from the volume of the empty chamber (air in the lungs is taken into consideration) -Body density and percent body fat are then calculated -Less cumbersome to administer -Takes only about 5 minutes
Physical Activity Questionnaires
-International Physical Activity Questionnaire (IPAQ) -Physical Activity Questionnaire for Adolescents (PAQ-A) -Leisure Time Exercise Questionnaire (LTEQ) STRENGTHS: -Inexpensive, allows for large sample sizes -Can be administered quickly and easily LIMITATIONS: -Reliability and validity problems associated with recall of activity, especially in children -Lack objectivity >Measures = Frequency, Intensity, Duration (and Energy Expenditure)
ANOVA (2-way)
-Like one-way ANOVA, but there are two independent factors -Used when we want to learn about the main effects of each factor individually, but also want to understand how they interact -a.k.a. Factorial ANOVA
Accelerometers
-Measure accelerations of the human body -Record activity patterns over a period of time -Benefits: small, non-invasive, large storage capacity, can be used widely in field settings -Different types: >Actigraph GT3X+ >ActivPAL >FitBit >Jawbone >Nike Fuelband STRENGTHS: -Measures actual movement -Valid & reliable in lab & field -Determine intensity of PA continuously in real-time -Easy & quick for data collection & analyses -Provide real-time data to personal electronic devices LIMITATIONS: -Sensor location may involve less movement in certain activities -Cost (large samples) -Reactive
Direct Observation
-Measure behavioral aspects of physical activity -Provides information on specific activities occurring in a variety of settings over time -Quantitative & qualitative information -Useful with younger children -Requires trained observers -Pen and paper instruments STRENGTHS: -Detailed quantitative & qualitative data -Describes PA in real time -Low financial cost -Computer software allows for immediate recording & analysis of data LIMITATIONS: -Time-intensive training -Time & labour intensive data collection -Limited sample sizes Observer presence (reactivity) -Limited validation against physiological criteria
Heart Rate Monitors
-Measures cardiorespiratory response to physical activity -Transmitter and belt worn around the upper body -Data commonly displayed on a wrist receiver -Downloaded via interface for analysis STRENGTHS: -Measures physiological relationship with energy expenditure -Valid & reliable in lab & field -Describes intensity of PA -Easy & quick for data collection & analyses LIMITATIONS: -Cost (large samples) -Data attrition -Discomfort over long periods -Age, sex, training status affect HR -No information on context of physical activity
Health Status (of a population)
-Morbidity >Incidence = >Prevalence = ...
Health Status (of a population)
-Mortality >Death rate = >Average lifespan = # of years one is expected to live
BMI
-Most common measure of body composition -Incorporates height and weight to estimate critical fat values at which disease risk increases BMI = Body Mass / Stature^2 Always expressed in metric units (kg/m^2 ) EXAMPLE: Body Weight = 172 lbs and Stature = 5'7" 1.Convert lbs to kg [2.2 lb = 1 kg] -Body Mass = 172 lbs = 78.2 kg 2.Convert in to m [39.37 in = 1 m] -Stature = 5'7" = 67 in = 1.70 m BMI = 78.2 / (1.70)2 BMI = 27
Nonexercise Estimates
-Nonexercise estimates -Uses age, measures of body composition and physical activity to estimate VO2max E.g., for women: VO2max = 44.310 - 0.326(age) - 0.227(BMI) + 4.471(SRPA) - 0.135(BMI x SRPA)
Oxygen Consumption
-O2 capacity/consumption plateaus (VO2 max) -VO2 VO2 max
Waist-to-Hip Ratio
-Predicts disease risk according to distribution of adiposity ("apple" vs. "pear" shape) -better than waist circumference DISEASE RISH ACCORDING TO WHR ...
Waist Circumference
-Predicts disease risk according to region of adiposity (waist versus other areas) DISEASE RISK ACCORDING TO WC ...
Aerobic Power Output
-Rate of mechanical work used to define exercise intensity -Usually expressed in Watts -Cycle ergometers (arm or leg) -Treadmills
Girth Measurements
-Requires a standard measuring tape -WOMEN: Upper arm, hip, and wrist measurements (cm) -Look up constants based on measurements and determine body density and percent fat according to predicting equations MEN: Waist and wrist measurements (inches) -Use table for percent body fat estimate
T-test (DEPENDENT means)
-Requires paired data -2 sets using the same measurement scale EXAMPLE: 1.BMI is measured b/4 and after a special diet program 2.Grip strength is measured in the dominant hand and the non-dominant hand -Pre-test and a post-test (each score in group A is paired with a score in group B) -T-test is based on the difference values (D) between the (pre- and post-test) -YOU NEED THE FOLLOWING: 1. n: number of pairs of measurements 2. D: the sum of all the differences 3. D : the sums of all the squares of all the differences STEP 1: calculate the t-value for DEPENDENT means t= STEP 2: after you know the t-value, you must determine the DEGREES OF FREEDOM (df) df = n - 1 = numbers of pairs - 1 STEP 3: determine the CRITICAL VALUE of t (Use table B.2) -Critical Value with an alpha = 0.05 (one-tailed and two-tailed) ==> t STEP 4: compare you t-value to the critical value STEP 5: make a decision -Reject H : upretest = u posttest -Do not reject H ex. t-values is > t-critical --> reject H
24-Hour Recall
-Requires trained interviewer -Interviewer helps subject remember all foods consumed and assists in determination of portion sizes -Common technique: -Interviewer asks what subject first ate and drank upon waking THAT day, and works forward to time of interview -Interviewer then begins at point exactly 24 hours ago and works forward to time of waking STRENGTHS: -Requires only 20-30 minutes -Fairly inexpensive -Provides detailed quantitative and qualitative data -Low subject burden -Low reactivity LIMITATIONS: -Time-intensive training -Time & labour intensive data collection -Seldom representative of usual intake Under/over reporting can occur -Requires cooperative subject; data can be withheld or altered
Bioelectrical Impedence
-Simple to administer, but accuracy is questionable -Sensors are applied to the skin and a weak electrical current is run through the body to estimate body fat, lean body mass, and body water -Based on the principle that fat tissue is a less-efficient conductor of an electrical current -The easier the conductance, the leaner the individual -Body weight scales with special sensors on the surface may also be used to perform this procedure -Hydration status can influence results ==> hydrated --> lean body --> high conductivity ==> dehydrated --> reads lean body fat as fat tissue --> low conductivity
Unweighed/Estimated Food Record
-Subject cooperation and literacy skills required -Less invasive - tends to be more acceptable to subjects -Some accuracy lost, but often balanced by better subject participation and adherence ESTIMATING FOOD PORTION SIZES -Significant problem with interpretation of portion sizes -Subjects should give as accurate description as possible, possibly by comparing to other objects (i.e. - deck of cards) -Food models or pictorial representations can also be used -Subjects can also be taught how to record net weight or volume on food labels and report proportion of the food that was consumed
Food Record/Diary
-Subject records, at time of consumption, the types and amounts of food and beverages consumed for a period of time ranging from 1-7 days -Weighed food record -Food and beverages weighed -Preferred method - considered more accurate -Unweighed/estimated food record -Portions estimated using household measures such as cups, teaspoons, tablespoons, etc. -PROSPECTIVE (looking forward in time) STRENGTHS: -Does not rely on memory -Can provide detailed data on intake and eating habits -Multiple day data more representative of usual intake -Valid up to 5 days LIMITATIONS: -Requires high degree of cooperation -Subject must be motivated and have average literacy level -Subjects willing to complete food records may not be typical -Quality of record declines in relation to the number of days studied -Reactive
Duplicate Food Collection (FOOD BUCKET)
-Subjects collect identical portions of all food and beverages consumed over a specified period of time -Duplicate portions are then chemically analyzed for nutrient content STRENGTHS: -Avoids coding errors and errors involved in the use of food composition tables -Can be used for nutrients and contaminants found in very small amounts for which food composition databases are incomplete LIMITATIONS: -Very expensive -Considerable subject burden -May underestimate intake
Aerobic Fitness
-The ability of the heart, lungs and blood vessels to supply oxygen to the working muscles -The ability of the muscles to use the available oxygen -A major component of fitness! -There is a significants relationship b/t aerobic fitness and mortality rates -O2 --> Lungs Heart Working Muscle -->kJ -muscles create mechanical energy measured in kJ How to measure aerobic fitness? 1.Exercise testing -Oxygen consumption (input to the body) -Power output (@ a maximum aerobic effort) -Field tests -12-minute run/walk -Generalized 1-mile run/walk -Shuttle run 2.Non-exercise approximations
Metabolic Equivalents (METs)
-The amount of oxygen used at rest -1 MET = 3.5 ml/kg*min (ml*kg^-1*min^-1)
Body Composition
-The different components that make up the human body - Fat - Muscle - Bone -Important in assessing health status and disease risk
EXAMPLE (Health)
-There are about 50,000 new cases of Parkinson's Disease each year in the US. -Incidence = 50,000 / adult population of US = 50,000 / 230,000,000 = 0.000217 -About 2 new cases in every 10,000 adults per year
Correlation Studies
-We know how to calculate the linear correlation coefficient, r -Testing the significance of a correlation: >Determine critical values of r (Table B4 in book) >Calculate p-values of r (Excel won't do this! Need better stats software) RESEARCH HYPOTHESIS: H : r > 0 H : r < 0 H : r ≠ 0 NULL HYPOTHESIS: H : = 0 DEGREES OF FREEDOM: df = total number of data pairs - 2 = n - 2
Interaction Effects (2-way)
-When the effect of one factor depends on another factor EXAMPLE: >There is an interaction between IRRIGATION and FERTILIZER on TREE GROWTH
ANOVA (2-way)
-When the variance WITHIN groups decreases, the variance BETWEEN groups becomes more apparent. F= -F increases -When the variance BETWEEN groups decreases, the variance WITHIN groups becomes more apparent. F= -F decreases
Main Effects (2-way)
-When there is a significant difference between different levels of a factor EXAMPLE: >There is a main effect of DIET on BMI >There is a main effect of AGE on BMI >There is a main effect of GENDER on STATURE
Hydrostatic Weighing (DUNK TANK)
-underwater weighing -most common technique used for decades -a person's "land" weight is compared w/ underwater weight -fat is more buoyant than lean tissue -most other indirect techniques have been validated against hydrostatic weighing (MORE BODY FAT==> LOWER DENSITY) CONS: -you have RESIDUAL LUNG VOLUME (air in your lungs still left after exhaling most of the air from your lungs) -some people are scared of being submerged under H2O -time consuming -not feasible to test large # of people -requires measurement of residual lung volume (if unknown, can be estimated) -difficult to perform on the aqua phobic
Table
...
Physical activity Exercise
-"Any bodily movement produced by the skeletal muscles that results in energy expenditure." -" A sub-category of physical activity, that is planned structured purposeful and repetitive."
Dietary Assessment - Methods
-24 hour recall -Food record/diary -Weighed (uses a scale) -Unweighed -Food frequency questionnaires -Diet history -Duplicate diet collection
Measuring Physical Activity
-30+ different instruments and/or methods -Pros and cons of each, depending on what information you are trying to capture -Characteristics assessed: >Frequency >Intensity >Duration
Factors (2-way)
-A FACTOR is a variable that separates data into groups. EXAMPLE: >Gender: MALE or FEMALE >Diet: HIGH-FAT or LOW-FAT >Treatment: PLACEBO, LOW DOSE, HIGH DOSE >Age: YOUNG, MIDDLE-AGED, ELDERLY >Physical Activity Level: SEDENTARY, ACTIVE
T-test
-A hypothesis that is used to determine if there is a significant difference b/t 2 groups -This is called the t-test for INDEPENDENT means -The t-test can also be used when there is only 1 group of subjects, and they are tested under 2 conditions -This is called the t-test for DEPENDENT means
ANOVA (ALL)
-All ANOVA tests concern the right-tail of the F-distribution -A large value of F represents a low probability that the data could have resulted if H0 is true. -Critical values of F are calculated based on degrees of freedom (size of dataset)
Food Frequency Questionnaires
-Assesses energy and/or nutrient intake by determining frequency of consumption of a limited number of foods known to be major sources of the dietary component under investigation -Lists approximately 100 food or food groups -May ask for indication of portion size -Portion numbers and sizes are statistically analyzed to estimate nutrient intake STRENGTHS: -Can be self-administered -Scannable computer scoring available -Inexpensive -May be more representative of habitual intake -Allows for examination of diet-disease relationships LIMITATIONS: -May not represent typical food or portion sizes consumed -Relies on ability of subject to describe diet
Diet History
-Assesses subject's usual dietary intake over an extended period of time -Collect general information on subject's health habits -Question subject about usual eating patterns -Verify accuracy of data provided -Subject also completes d-day food record STRENGTHS: -Assesses usual nutrient intake -Can detect seasonal changes -Data on range of nutrients can be collected -Correlates well with biochemical markers of intake (i.e. - protein intake and nitrogen excretion) LIMITATIONS: -Time consuming and expensive -Labor intensive - requires highly trained interviewers and complex coding systems -May overestimate nutrient intake -Requires cooperative subject with ability to recall usual diet
Skinfold Thickness
-Based on the principle that approximately half of the body's fat tissue is directly beneath skin -Reliable measurements can give a good indication of percent body fat -Skinfold test is done with pressure calipers -Several sites are measured and percent fat is estimated from the sum of the three sites -All measurements should be taken on the right side of the body ANATOMICAL LANDMARKS FOR SKINFOLD MEASURMENTS -Chest -Abdomen -Triceps -Suprailium -Thigh %BF -Females S=triceps + suprailium + thigh (in mm) BD=1.099421 - (0.0009928*S) + (0.00000023*S^2) - (0.0001382*Age) %BF=(495/BD) - 450
Body Composition Changes for Adults in the United States
-Because of the typical reduction in physical activity, each year the average person gains 1.5 lbs of body fat and loses a half a pound of lean tissue
Physical Activity Logs (PALS)
-Bouchard Physical Activity Record >Measures = Frequency, Intensity, Duration (and Energy Expenditure)
Summary
-Choice of method to assess dietary intake depends on underlying purpose of assessment -Dietary assessment can be retrospective or prospective -Each method had strengths and limitations -Biomarkers can be used as measures of the validity of the dietary intake data collected
Summary(2)
-Choice of method to assess physical activity depends on underlying purpose of assessment -PA assessment can be retrospective or prospective -Each method had strengths and limitations -Markers of energy expenditure can be used as measures of the validity of the physical activity data collected
Measurement of Health Status
-Definition of Health: 1. Absence of: >Physical pain >Physical disability >Conditions likely to cause death 2. Emotional well-being 3. Satisfactory social functioning
Assessing Body Composition
-Direct Techniques --> involve looking inside the body and measuring the different tissues by volume (or area) --> EXAMPLES: (time duration ~10-15 minutes for all) > Dual energy X-ray absorptiometry (DXA) > Magnetic Resonance Imaging (MRI) > Computed Tomography (CT)
ANOVA
ANALYSIS OF VARIANCE -a method for testing the hypothesis that 3 or more population means are equal EXAMPLE: H : µ1 = µ2 = µ3 = . . . µk H : At least one mean is different ASSUMPTIONS: -Each group is comprised of a randomly selected sample -Scores in each group are normally distributed -The variances of each group population are homogenous TEST STATISTIC: -F-score -The F- distribution is not symmetric; it is skewed positively. -The values of F can be 0 or positive; they cannot be negative. -The F-distribution changes shape with respect to degrees of freedom ADD + skewed example -F-distribution ...
Health Status (of a population)
Disability-Adjusted Life Years (DALY) -a measure of overall disease burden, expressed as the cumulative number of years lost due to ill-health, disability or early death >DALY = YLD + YLL -YLD (years lived with disability) -YLL (years of life lost)
F Statistic (1-way)
F = MSBetween MSWithin -If there is no effect of treatment, then MSbetween will be relatively small and F will be small. -If there is a significant effect of treatment, then MSbetween will be large relative to MSwithin and F will be large -If F > Fcrit, then there is an effect of treatment, and we can reject the null hypothesis CALCULATING F: >k = number of groups >N = total number of data points >X = the mean value of each group then we calculate... >SStotal = Sum of squares of all data >SSw = Sum of squares within groups >SSb = Sum of squares between groups >df = degrees of freedom -MSw = Mean sum of squares within groups -MSb = mean sum of squares between groups = F (value) Use an ANOVA Table: .... -We have our = F (value) then we determine the critical value of F for alpha = 0.05 (use table B.3) >You need: 1.df for the Denominator (MSw) 2.df for the Numerator (MSb) 3.Type I Error rate -test statistic (F) is > F-critical ==> reject null hypothesis (at least one of the groups is different than the others (research hypothesis))
Health and Fitness
Fitness --> how well a living-being thrives in its environment -Health-Related Physical Fitness >Body composition -Heart disease, diabetes... >Aerobic fitness -Heart disease, ... >Flexibility -Low back pain >Muscular endurance -Chronic fatigue >Muscular strength -Function, independence
EXAMPLE (Health)
In Division 1 Collegiate football players, the prevalence of: >Obesity is 21% >Insulin resistance is 21% >Metabolic syndrome is 9% If there are 5800 Div-1 college football players, how many have insulin resistance? >X = 5800 * 0.21 = 1218
REVIEW
In an experimental study, 30 soldiers are randomly assigned to 4 groups. Each group receives a different style of training. The researchers want to know if the training style has an effect on the soldiers' physical fitness test scores. What is the critical value for F (let a = 0.05): 1. -2.69 2. 0 3. 2.98 4. There is not enough information!
Intensity
METs: multiples of the metabolic equivalent of sitting quietly for one hour >Sedentary < 1.5 METs >Light 1.5 < 3.0 METs >Moderate 3 - 6 METs >Vigorous > 6 METs
Physical Activity Level (PAL)
PAL = .... Add Table
DXA
PROS: -Dual Energy X-Ray Absorptiometry (DXA) -Most frequently used by research and medical facilities -Considered the "GOLD STANDARD TECHNIQUE" for body composition assessment -Uses low-dose beams of x-ray energy -Measure FAT MASS, FAT FREE MASS and BONE DENSITY -Procedure is simple --> takes only 15 minutes to administer -Safe --> no problem w/ radiation exposure,etc. CONS: -not portable -very expensive ($) -cannot be used on an individual with metal implants
MRI
PROS: -Magnetic Resonance Imaging (MRI) -Used primarily by research and medical facilities -Uses high powered magnets and imaging software -2nd "GOLD STANDARD TECHNIQUE" -Measures TOTAL and REGIONAL ADIPOSITY -Safe --> no problem with radiation exposure, etc. CONS: -MORE expensive than DXA ($$) -If you are claustrophobic... you might not be able to get and MRI because of how tight spaced the machine is -cannot be used on an individual with metal implants
CT
PROS: -Computed Tomography (CT) -Measure regional adiposity by imaging multiple cross- sectional slices -Measures fat layers within the body cavity, limbs or organs -Sequential images are segmented into adipose and non-adipose tissue pixel values measured in HOUNDSFIELD UNITS (HU) -Procedure is simple, time required depends on # of slices taken CONS: -Uses higher-dose X-ray energy -Radiation exposure can be a drawback -MORE EXPENSIVE than DXA & MRI ($$$)
Body Composition
Percent Body Fat (%BF) --> wt of fat in the body relative to the person's total wt. - Essential Fat: Minimal amount of body fat needed for normal physiological functions; constitutes ~3-5% of total wt. MEN ~8-12% WOMEN - Storage Fat: Body fat in excess of essential fat; stored in adipose tissue Lean Body Mass --> Body wt. without body fat
Correlation Studies
Step 1: Calculate test statistic, r Step 2: Look up the critical value of r Step 3: Compare your r to r Step 4: Reject the null hypothesis? -using a 2-tailed test w/ alpha=0.05
Methods to Assess PA
Subjective Measures: >Physical activity logs (PALS) >24 hour recall >7 day recall >Previous month recall >Previous year recall >Questionnaires >All are self-administered or interviewer administered, either in-person or on telephone Objective Measures: >Accelerometers -Marker of movement >Heart rate monitoring -Physiological response to PA >Pedometers -Number of steps taken >Direct observation -Observed or videotaped >GPS systems -Newer application; measures population movement
ANOVA (1-way)
Test Statistic for One-Way ANOVA F= variance BETWEEN groups/ variance WITHIN groups -An excessively LARGE F test statistic is evidence against equal population means. ASSUMPTIONS (same as b/4): -Group populations are normally distributed -Group variances are equal ANOVA asks: -Judging from our samples, what is the probability that they all came from the same population? ANOVA breaks the total variance, s , into 2 parts: 1.Variance BETWEEN groups 2.Variance WITHIN groups VARIANCE: -Population variance: o = -Sample mean: x = .... ....
REVIEW
The variation BETWEEN groups represents: 1.How much the individual data vary with respect to their group means 2.How much the group means vary with one another 3.How much the individual data vary with respect to the mean value of all data 4.The denominator of the F-statistic
REVIEW
There is no correlation between aerobic fitness and grip strength. What is the probability of measuring r > 0.7293 within a group of 6 randomly selected people? 1. 0% 2. 2.5% 3. 5% 4. 10% 5. 95%
REVIEW
Which statement is true about the F-distribution? 1.It is negatively skewed 0 ≤ F ≤ 1 2.It can be approximated by the normal curve 3.It is used in analysis of variance 4.The standard deviation is always 1