ENS 305 Unit One

Formative testing

Formative testing refers to assessments conducted during the learning process to monitor student understanding and inform instruction, providing feedback that can help improve student performance (week-to-week, day-to-day)

Summative testing

Summative testing refers to assessments administered at the end of an instructional period to evaluate and measure student learning against standards, often resulting in a final grade (

Norm-referenced standard

used to compare a student's performance against a group, providing insight into relative standing.

Criterion-referenced standard

used to measure a student's performance against predefined criteria or standards, indicating whether a student has mastered specific skills or knowledge.

Cognitive domain of testing

refers to the mental skills and knowledge involved in learning, encompassing processes such as thinking, reasoning, and problem-solving.

Affective domain of testing

pertains to emotions, attitudes, and values related to learning, focusing on aspects such as motivation, interest, and feelings.

Psychomotor domain of testing

focuses on physical movement, coordination, and motor skills

What are the four facets of an effective hypothesis?

Population, condition, variable, direction

Retrospective study design

Reliability

Consistency and repeatability of a measurement or test

Systematic error

Not utilizing equipment properly, produces variable results

Random error

Can come from weather, natural variations (time of day, hunger, mood, noise), imprecise measurement (different testers read analyzing tools differently), or individual differences (perception of pain)

Continuous data

Includes a potentially infinite number of values (fractions and decimals) [10.57 s, 1.68 m]

Discrete data

Limited to a specific number of values (whole numbers) [27 push-ups, jersey numbers, grade of hamstring strain]

Validity

the degree to which the conclusions drawn from a data analysis or statistical test are accurate, justified, and reflective of a true effect

Nominal

Categorizes data without any order or magnitude (male/female, jersey #)

Ordinal

Ranks data in order, but intervals between ranks are not equal (fitness classifications [excellent, good, fair, poor], pain rating scales [low, moderate, high])

Ratio

Has equal intervals and a true zero point (height, weight, body fat %, HR, VO2 max)

Interval

Has equal intervals between values, but no true zero point (scores on a test, temperature in C* or F*

Mean

Sum of all values divided by the # of values, used when data is normally distributed

Median

Middle score when the data is ordered; often used when outliers are present

Mode

The most recurring value of a data set; often used with nominal data variability

For nominal data, which descriptive statistic measure should be used?

Mode

For ordinal data, which descriptive statistic should be used?

Median or mode

For interval/ratio data, which descriptive statistic measure should be used?

Mean if no skew, median if there is a skew

What is the p value representative of?

This gives us the

percentage that the finding comes by chance (as a decimal). Typically is <0.05 in kinesiology.

Type I error

Rejection of the null hypothesis when it is true

Type II error

Not rejecting the null hypothesis when it is false

Positive skew

To the left of the mean

Negative skew

To the right of the mean

Mesokurtic

Curved an average amount

Platykurtic

Flat, negative kurtosis value

Leptokurtic

Peaked, positive kurtosis value

How do you calculate z-score with skewness/kurtosis?

Z-score= skewness(or kurtosis)/std. error

What does the 50th percentile represent?

The median

What does the 90th percentile represent?

Elite

What is a z-score?

Distance from the mean in standard deviations

Z-score formula

observation-mean/standard deviation

What will a z-score of +1 give you? (In percentile?)

High, close to elite

What will a z-score of -1 give you?

Below average

What will a z-score of 0 give you?

average

What does bias in testing lead to?

A threat to validity, results in type I errors

What do errors in testing lead to?

A threat to reliability, which results in type II errors

What are your assumptions for pearson’s R?

Two continuous variables, those two continuous variables are paired, your variables have a linear relationship, there are no significant outliers, and your variables are normally distributed