Term
| This type of study has DV, IV, and Levels |
|
Definition
|
|
Term
| The full detail of an IV; that to which is being directly compared; Every IV has at least 2 of these. |
|
Definition
|
|
Term
| If the IV is "sex", what are the levels? |
|
Definition
|
|
Term
| If the levels are hot pack and ultrasound then what is the IV? |
|
Definition
|
|
Term
| If you take Post Test minus Pre Test what is that? |
|
Definition
|
|
Term
| What type of study only has IV and DV, no levels |
|
Definition
|
|
Term
| So if you are measuring height, weight, and ability to jump, what are the DV, IV, and levels? |
|
Definition
DV - ability to jump
IV - height, weight
No levels |
|
|
Term
| What if you were trying to predict body fat by using calibers at three locations (supra-iliac, scapula, calf). What are the IV, DV, and Levels? |
|
Definition
DV - Body Fat
IV - Supra-iliac, scapula, and Calf
NO LEVELS |
|
|
Term
| The relationship or association between two sets of data. IV only, no DV or levels |
|
Definition
|
|
Term
| Give an example of a correlation study |
|
Definition
Height vs. Weight
Amount of exercise vs. heart rate |
|
|
Term
| Name the 3 steps to testing a hypothesis |
|
Definition
1. State the hypothesis
2. State the alpha level
3. Collect data and perform statistical analysis |
|
|
Term
| What are that parts to stating the hypothesis? |
|
Definition
IV, DV, Levels
If - Then
Null - most frequently used
-The idea is to not have a bias |
|
|
Term
|
Definition
| The chance you are willing to take to be wrong. The level of significance. The chance you are willing to take for being wrong with your statistical analysis. |
|
|
Term
| What does an alpha level of .05 mean? |
|
Definition
-5 in 100 (5%) of being wrong if accept/reject null hypothesis
-95 in 100 (95%) of being correct if accept/reject null hypothesis |
|
|
Term
| What does an alpha level of .01 mean? |
|
Definition
| You are only willing to take a 1% chance of being wrong. You are 99% you are correct. |
|
|
Term
| What is a common level for PT studies? |
|
Definition
|
|
Term
| If you accept the null hypothesis, is this a significant difference? |
|
Definition
|
|
Term
| If you reject the null hypothesis, is this a significant difference? |
|
Definition
|
|
Term
| The difference obtained in collecting data was due to chance; if you repeat the experiment, same thing may or may not occur. |
|
Definition
| NO Significants - Accept the null |
|
|
Term
| A significant difference exists between the groups. If the experiment was done again the same thing would happen every time. |
|
Definition
| Yes significant - Reject the null |
|
|
Term
| The difference is due to manipulation of variable, NOT due to chance. If you repeat the experiment, you will get the same result. Results can be generalized from sample to population. |
|
Definition
|
|
Term
| This exists if the probability is less than .05 |
|
Definition
|
|
Term
|
Definition
|
|
Term
| What is significant difference (3 bullets) |
|
Definition
1. different is due to manipulation of variable, not due to chance
2. if repeat experiment, will get same results
3. results can be generalized from sample to population |
|
|
Term
|
Definition
-Reject the null hypothesis, when null is really true
- historically most talked about
- You say there is a difference when there really was NOT |
|
|
Term
| Give an example of a type I error |
|
Definition
| Saying something is different and later studies indicate that no difference occurs |
|
|
Term
|
Definition
- Accept the null hypothesis, when a difference really exists
- Saying that there is no difference when there IS |
|
|
Term
| Give an example of a Type II error |
|
Definition
| Saying there is no difference and later studies indicate that a difference exists |
|
|
Term
| When are errors discovered? |
|
Definition
| When later studies are performed |
|
|
Term
| Quantitative Date (scores from many subjects) are: (4) |
|
Definition
Organized
Described
Analyzed
Interpreted |
|
|
Term
| Enables researchers to make generalities from data with increased assurance of being correct. Do NOT prove anything. Are a tool that assists in the probability of making the correct decision. |
|
Definition
|
|
Term
| Name 4 ways to determine the value of statistics |
|
Definition
1. Understand and interpret literature
2. Determine worth of test or instrument
3. Conduct research
4. Discriminate good and bad research |
|
|
Term
| Name 2 types of statistics |
|
Definition
General
Specific (based on levels of measurement) |
|
|
Term
| Name types of General statistics |
|
Definition
|
|
Term
| Describes a picture; Does NOT deal with a significance. Central tendency (mode, median, mean). Variability (range, variance, standard deviation). Normal Distribution |
|
Definition
|
|
Term
| Shows how spread out the scores are |
|
Definition
| Variability (range, variance, standard deviation) |
|
|
Term
| Deals with "significance". T-test, ANOVA, Correlation, other |
|
Definition
|
|
Term
| Name the 2 types of specific statistics |
|
Definition
Parametric
Non-parametric
Based on level of measurement |
|
|
Term
| Name the 4 levels of measurement |
|
Definition
Nominal
Ordinal
Interval
Ratio |
|
|
Term
| Numerical scale indicate a classification; inclusion in one category is not better/worse, larger/smaller, etc. than another |
|
Definition
|
|
Term
| Give a few examples of Nominal |
|
Definition
Male/Female
Number on athlete's uniform
Yes/No
Agree/Disagree
1=OA 2= RA 3=nl |
|
|
Term
| Order or rank; comparison of better than or greater than can be used. No proportionality to it |
|
Definition
|
|
Term
| Give a few examples of Ordinal |
|
Definition
MMT
Pain scale
Finish in a race (1st, 2nd, 3rd) |
|
|
Term
| Equal intervals between number, but not related to absolute zero; all "made up" scales. There IS proportionality to this |
|
Definition
|
|
Term
| Give a few examples of Interval |
|
Definition
IQ
Fahrenheit
Centigrade
Calendar Years
Gymnastic Score |
|
|
Term
| The highest level of measurement; contains all characteristics of other three scales; there is an absolute zero |
|
Definition
|
|
Term
| Give a few examples of Ratio |
|
Definition
ROM
Height
Weight
Force
Kelvin Temp |
|
|
Term
| Name the 2 types of specific statistics |
|
Definition
| Parametric and non-parametric |
|
|
Term
| More "power" (getting the correct answer without error); increased respect and sophistication; based on ratio and interval data; t-test, ANOVA, Pearsons correlation, ICC |
|
Definition
|
|
Term
| Simpler tests; increased chance of error; less sophistication and respect; based on nominal and ordinal data; Chi Square, Spearman Rho correlation, Sign test, and Median test |
|
Definition
|
|
Term
| These statistics describe the data of a group. They are a measure of central tendency and variability |
|
Definition
|
|
Term
| A single score that represents all scores in a distribution (mean,median,mode) |
|
Definition
|
|
Term
| The score that appears most frequently. There may be more than one. Frequently used with nominal data. |
|
Definition
|
|
Term
| The value above which there are as many scores as below it. Odd number of scores in the set of scores - the middle score. Even number of scores in the set of scores - the mean of the two middle numbers. Frequently used with ordinal data. |
|
Definition
|
|
Term
| The sum of a set of scores divided by the number of scores. Referred to frequently as the "average". Indicated by (line over x). Most frequently used with interval and ratio scale. |
|
Definition
|
|
Term
| Indicates how scores are spread in a distribution. Range, standard devaition, variance |
|
Definition
|
|
Term
| Represents spread of all scores in the distribution. Does not provide any info about how scores are distributed between high and low scores |
|
Definition
|
|
Term
| The difference between the largest and smallest score |
|
Definition
|
|
Term
| The list of the lowest and the highest score |
|
Definition
|
|
Term
| Most frequent measure of variability. Tells you how the scores are spread. Are they spread out or clumped together. |
|
Definition
|
|
Term
| What does 1 standard devation mean? |
|
Definition
| 68% of all scores are above and below measure of central tendency (mean) |
|
|
Term
| What does 2 standard deviations mean? |
|
Definition
| 95% of all scores are above and below measure of central tendency (mean) |
|
|
Term
| What does 3 standard deviations mean? |
|
Definition
|
|
Term
| If you have a mean of 50 and a standard deviation of 5. What would 1 SD mean and 2 SD mean and 3 SD mean? |
|
Definition
1 SD = 68% of scores are between 45 & 55
2 SD = 95% of scores are between 40 & 60
3 SD = 98% of scores are between 35 & 65 |
|
|
Term
| Very important with inferential statistics. ANOVA. Analyses how scores are spread around the mean. Not used in descriptive stats. Variance = SD (squared) |
|
Definition
|
|
Term
| Used to determine relationship (association) between two or more variables. Does not provide information about differences. No levels or DV, IV ONLY |
|
Definition
|
|
Term
|
Definition
Correlation Coefficient
Perfect Relationships |
|
|
Term
| If the correlation coefficient is positive what does that mean? |
|
Definition
Direct relationship
Ex: Increase in height means increase in weight |
|
|
Term
| If the correlation coefficient is negative what does that mean? |
|
Definition
Inverse relationship
Ex: Increase in weight, decrease in jump height |
|
|
Term
| What is a perfect correlation |
|
Definition
|
|
Term
| what is a zero correlation |
|
Definition
|
|
Term
| What is a weak/low correlation |
|
Definition
|
|
Term
| In physical therapy, to be reliable, the correlation coefficient should be __ or higher. |
|
Definition
|
|
Term
| Is correlation a cause and effect? |
|
Definition
|
|
Term
| A "family of coefficients" that allows analysis of two or more repeated measures. Used to analyze measurement reliability. |
|
Definition
| ICC (Intraclass correlation) |
|
|
Term
|
Definition
Pearsons (2 variables)
-Interval/Ratio |
|
|
Term
| Non-Parametric Correlation |
|
Definition
Spearman Rho (2 variables)
- test 1 vs. test 3
- isokinetic vs. vertical jump
- self esteem vs. motivation |
|
|
Term
| Name the 7 steps to statistical testing of differences (CALCULATOR AND TABLE) |
|
Definition
1. State Hypothesis
2. Determine alpha level
3. Determine if sample is independent/dependent
4. Use formula to calculate test statistic
5. Calculate degrees of freedom
6. Determine "critical" value of test statistic given df and alpha level
7. Compare "critical" value to obtain test statistic |
|
|
Term
| Give an example of an independent and a dependent variable. |
|
Definition
Indep = male/female
Dep = pre & post test |
|
|
Term
| A way of correcting for possible erros |
|
Definition
|
|
Term
| The number you have to beat for performing your statistic |
|
Definition
|
|
Term
| If the test statistic is greater than the critical value then |
|
Definition
| there is a significant difference which is what we want |
|
|
Term
| Name the 6 steps to statistical testing of differences (COMPUTER) |
|
Definition
1. State hypothesis
2. Determine alpha level
3. Determine is sample is indep/dep
4. Use appropriate program to calculate test statistic
5. Computer calculates probability
6. Compare obtained probability with alpha level |
|
|
Term
| If obtained probability is less than the alpha level then |
|
Definition
| there is a significant difference which is what we want |
|
|
Term
| Comparison of TWO groups. Differences between group means divided by mean square error. The simplest statistic. |
|
Definition
|
|
Term
| You want the t-value to be big. To get that MSE should be small. X1 shoudl be a lot bigger than X2 |
|
Definition
|
|
Term
|
Definition
|
|
Term
| If t is __ there is a better chance of significant |
|
Definition
|
|
Term
| Large differences in group mean = |
|
Definition
|
|
Term
|
Definition
|
|
Term
| Small difference in group mean or large error |
|
Definition
| no significant difference |
|
|
Term
-Represented in denominator of formula
-Small sample size
-Difference btw subjects in same group
-Difference in accuracy of measurement
-Difference in treatment of subjects |
|
Definition
|
|
Term
| A small sample size will make __ smaller and the chance for error ___ |
|
Definition
|
|
