Lecture 2- Variables and Measurements

How should dependent variable be defined?

Operational Definition!

Operational definitions provide a precise description of how an

investigator measured (or manipulated) the dependent variable

Operational definitions increase the odds that investigators in different laboratories can make the same measurements

How would you describe weight operationally?

You would describe how the subjects were weighed (e.g., the make and the model of the scale) and the degree of precision (e.g., to the nearest 0.1 or 0.01g)

How would you define birthday cake operationally?

Provide the recipe

Provide an operational definition for blood pressure. (3 points)

Using a commercial pressure cuff to measure systolic and diastolic pressure. (Method, device, precision)

What are the two main types of error?

Bias is the average error in judgment (systematic deviation), while noise is the variability or inconsistency in those judgments (random scatter)

Name some sources of noise and bias in a 1950s bathroom scale

Noise:

• Person moving or shifting weight on the scale

• Inconsistent positioning of feet

• Uneven floor surface

• Difficulty reading the dial precisely (small markings)

Bias:

• Scale is miscalibrated (always reads too high or low)

• Zero not set correctly before use

• Worn or damaged internal springs/mechanism

• Dial consistently offset from true value

From noise and accraucy, which one is easier to measure?

Noise is easier to measure, accuracy is harder to quantify.

Describe how bias and noise would be represented graphically?

Bias is the deviation from that broken line

What are sources of bias in measurement?

Observer bias: Occurs when a researcher's expectations, personal prejudices, or subjective opinions unintentionally influence how they observe, record, or interpret data in a study.

Subject Bias: (a.k.a. Observer or Hawthorne effect): Occurs when participants in a study consciously or unconsciously alter their behavior or responses to match what they believe the researcher expects, rather than acting naturally.

Problems with measurment device

How can you minmize observer bias?

Automate data collection

Simplify data collection

Blind the observer

How do you minimize subject bias?

Conduct study in natural setting

Blind the participant

Sneakily take measurements without them noticing

How would you minimize noise from a measurement device?

• Make sure your measuring device is functioning normally

• Make sure the operator of the device is well trained and rested

• Make sure you use a measuring device that is appropriate to the measurements that you are making

(e.g., a scale that measures human weight accurately probably will notmeasure mouse weight accurately)

What are the different scales of measurement?

Nominal- Comparing groups qualitatively (skin color, gender) (no hierarchy in ranking)

Ordinal- Numbers represent ranking (hierarchy) with no true zero eg, professor ranking

Interval scale numbers- Ranked order with constant interval size, no true o, eg. temperature (ratio statement not allowed)

Ratio scales- Zero value representing absence of the property being measured (height, weight temperature) (ratio statatment allowed)

Name the most appropriate measurements for each dependent measure?

Course evaluations

Degree of aversion to a food

Body mass

Gender

Time

Senior class ranking

Temperature in degrees Kelvin

Distance jumped

Type of vehicle

Course evaluations- Ordinal

Degree of aversion to a food- ordinal

Body mass- ratio

Gender- nominal

Time- interval

Senior class ranking- ordinal

Temperature in degrees Kelvin-ratio

Distance jumped- ratio

Type of vehicle-nominal

B

A

D