Title | Psych 203 Notes #4 - Professor Gregory Speth Measurement: Operational Definitions, Measurement Error, |
---|---|
Author | Erin Aduna |
Course | Introduction To Psychological Research. |
Institution | Montclair State University |
Pages | 3 |
File Size | 76.1 KB |
File Type | |
Total Downloads | 15 |
Total Views | 156 |
Professor Gregory Speth
Measurement: Operational Definitions, Measurement Error, Reliability, Validity
Scales of Measurement: NOIR= Nominal, Ordinal, Interval, Ratio
Descriptive Statistics: CT Central Tendency
Median, Mode, Mean...
PSYCH 203 Montclair State University Spring 2017 Notes #4 _____________________________________________________________________________
MEASUREMENT: Operational Definitions: specific as possible. Examples may include frequency, intensity, duration. Measurement Error: difference between sample statistic and population parameter. It can also refer to any measurement error. Always present but all attempts to minimize it are made. Reliability: one of the two most important constructs in measurement. How consistent a measurement is. There are many forms of reliability. Established before validity. Validity: one of the two most important constructs in measurement. How well a test/measure assesses what it purports to measure. There are many forms of validity.
SCALES OF MEASUREMENT: NOIR: hierarchy of measurement. Nominal: difference. Ordinal: rank/order. Interval: (AKA scale data, AKA score data) equal intervals. Ratio: (AKA scale data, AKA score data) absolute zero, “measurement ideal.” ______________________________________________________________________________
DESCRIPTIVE STATISTICS: Descriptive Statistics: used to, as the name implies, Describes a set of data. Examples of Descriptive Statistics include skewness, CT, SD, Variance, the presence of outliers, etc. Central Tendency (CT): a statistic used to describe or represent a set of data. A number, or group of numbers, used to describe where the bulk of the data points are. There are many different CT statistics used (i.e., 25+), most statistics use the mean, mode, or median. The mean is the most commonly used but, as with most statistics, the choice of which CT statistic(s) to use
typically depends on the nature of the investigation, nature of the data, and the type of variable being measured (i.e., scale of measurement). Mean: represents an “average” score (� X/N). Mode: the most common score. Median: the middle data point (i.e., the 50th percentile). Variability: a descriptive statistic. Typically a single number used to describe how spread out a set of data is. The standard deviation is the most commonly used but, as with most statistics, the choice of which CT statistic(s) to use typically depends on the nature of the investigation, nature of the data, and the type of variable being measured (i.e., scale of measurement). Range: highest and lowest number. Variance: a statistic used to measure the total spread of a distribution, measures how heterogeneous (dissimilar) a set of data is, it is the square of the standard deviation. Standard Deviation (SD): as the name implies, measures the spread of a set of data in Standard units, thus it can be compared with other data sets. SD is used to describe how individual scores differ from the central tendency and other scores. Standard Deviation represents “average variability.” AKA: Variability, Dispersion. SD is the square root of the variance. ______________________________________________________________________________
INFERENTIAL STATISTICS: Inferential Statistics: As the name implies, a statistical process in which a Sample of a Population is measured and a hypothesis or INFERENCE is made from Sample to the larger Population. It is the basis for almost all statistical/quantitative research. For example, suppose a researcher wanted to measure math aptitude among college freshman in the U.S. He or she would test a sample group (e.g., n=250) from a local college. Based on the results, the researcher would then INFER that the Sample is representative of the entire Population (college freshman in the US) and draw conclusions regarding the entire Population. Obviously, inferential statistics have serious limitations, one must be cautious about generalizing from a sample to a whole population. If the Sample does not accurately represent the entire population on the variable being measured then inferences cannot, and should not, be drawn regarding the larger Population. Population: entire group in question. Parameters: variables/characteristics within the population.
Sample: part/section of the population studied used to make inference about the population. Statistics: variables/characteristics within the sample. ______________________________________________________________________________...