BODY COMPOSITION EXAM WITH
VERIFIED SOLUTIONS
Body density - Correct Answers -total body mass expressed relative to total body
volume; varies from person to person because the percentage of components vary;
reference man was 1.064
Fat-free body (FFB) - Correct Answers -all residual lipid-free chemicals and tissues
including water, muscle, bone, connective tissue, and internal organs
Lean body mass - Correct Answers -fat-free mass plus essential lipids
Correlation - Correct Answers -measures the strength of association or degree of the
relationship between two variables; how well two things relate to each other; strong =
0.8 or greater; r = 0-1; + or - is direction only; generally linear data
Coefficient of determination - Correct Answers -the square of the correlation coefficient
and represents the amount of variance shared by variables X and Y; N% of the
variability is explained by Y and 100-N% is unexplained; r^2
Regression - Correct Answers -used to predict one variable from one or more other
variables; aka prediction
Bivariate regression - Correct Answers -predicting one thing from something else; must
know the relationship between the two variables; on a graph, find a line of best fit
Regression line - Correct Answers -the line that minimizes the squared deviations from
the line
Multiple regression - Correct Answers -predicting one thing from many variables
Polynomial regression - Correct Answers -one of the variables has a square; puts a
curve in the line of best fit
Basics of regression analysis - Correct Answers -Y=a+bX where Y = dependent
(criterion) variable; X = independent (predictor) variable variable; a = y-intercept, b =
slope
, T-test - Correct Answers -tests the difference between the means; Paired = same group
tested in all tests (want small SEE, TE and non-significant difference); Independent =
different group of people are tested on different tests
Weight loss diseases - Correct Answers -Heart failure
increased TBW & ECF
Cachexia
COPD
decreased FFM & BMC
Cystic fibrosis
decreased FM, BMC, BMD
Type I diabetes
Hyperthyroidism (Grave's disease)
decreased FM & FFM
Cachexia (ca-say-she-a) - Correct Answers -Muscle wasting - due to disease, not old
age
Body cell mass: better indicator of survival than weight - using phase angle from BIA
decreased phase angle (<5.6) shows a lower survival rate
Cachexia diseases - Correct Answers -Anorexia nervosa
HIV/AIDS
Kidney failure/dialysis
Cirrhosis/liver disease
Spinal cord injury
MS & MD
Standard Error of Estimate - Correct Answers -on average, the distance that people are
away from the regression line (line of best fit); smaller is better, because then people
are closer to that line
Total error - Correct Answers -how far away people are from the line of identity; smaller
is better; will generally be larger
Line of Identity - Correct Answers -slope = 1 and y-intercept = 0; if the data points are
on or very close, then the prediction is very good
Cross-validation - Correct Answers -use a smaller group of people and repeat the study
to see if the new equation (prediction) works again
Criteria for Evaluating Prediction Equations - Correct Answers -1. Reference method?
multiple component model
2.Sample size? minimum of 20 people per variable
3.Applicability of equation (general or population-specific) depends on the population
and intent
4.Cross-validated? yes
VERIFIED SOLUTIONS
Body density - Correct Answers -total body mass expressed relative to total body
volume; varies from person to person because the percentage of components vary;
reference man was 1.064
Fat-free body (FFB) - Correct Answers -all residual lipid-free chemicals and tissues
including water, muscle, bone, connective tissue, and internal organs
Lean body mass - Correct Answers -fat-free mass plus essential lipids
Correlation - Correct Answers -measures the strength of association or degree of the
relationship between two variables; how well two things relate to each other; strong =
0.8 or greater; r = 0-1; + or - is direction only; generally linear data
Coefficient of determination - Correct Answers -the square of the correlation coefficient
and represents the amount of variance shared by variables X and Y; N% of the
variability is explained by Y and 100-N% is unexplained; r^2
Regression - Correct Answers -used to predict one variable from one or more other
variables; aka prediction
Bivariate regression - Correct Answers -predicting one thing from something else; must
know the relationship between the two variables; on a graph, find a line of best fit
Regression line - Correct Answers -the line that minimizes the squared deviations from
the line
Multiple regression - Correct Answers -predicting one thing from many variables
Polynomial regression - Correct Answers -one of the variables has a square; puts a
curve in the line of best fit
Basics of regression analysis - Correct Answers -Y=a+bX where Y = dependent
(criterion) variable; X = independent (predictor) variable variable; a = y-intercept, b =
slope
, T-test - Correct Answers -tests the difference between the means; Paired = same group
tested in all tests (want small SEE, TE and non-significant difference); Independent =
different group of people are tested on different tests
Weight loss diseases - Correct Answers -Heart failure
increased TBW & ECF
Cachexia
COPD
decreased FFM & BMC
Cystic fibrosis
decreased FM, BMC, BMD
Type I diabetes
Hyperthyroidism (Grave's disease)
decreased FM & FFM
Cachexia (ca-say-she-a) - Correct Answers -Muscle wasting - due to disease, not old
age
Body cell mass: better indicator of survival than weight - using phase angle from BIA
decreased phase angle (<5.6) shows a lower survival rate
Cachexia diseases - Correct Answers -Anorexia nervosa
HIV/AIDS
Kidney failure/dialysis
Cirrhosis/liver disease
Spinal cord injury
MS & MD
Standard Error of Estimate - Correct Answers -on average, the distance that people are
away from the regression line (line of best fit); smaller is better, because then people
are closer to that line
Total error - Correct Answers -how far away people are from the line of identity; smaller
is better; will generally be larger
Line of Identity - Correct Answers -slope = 1 and y-intercept = 0; if the data points are
on or very close, then the prediction is very good
Cross-validation - Correct Answers -use a smaller group of people and repeat the study
to see if the new equation (prediction) works again
Criteria for Evaluating Prediction Equations - Correct Answers -1. Reference method?
multiple component model
2.Sample size? minimum of 20 people per variable
3.Applicability of equation (general or population-specific) depends on the population
and intent
4.Cross-validated? yes
