Research Application of Psych Quiz #3

What is the purpose of a regression analysis?

predict outcomes, test the strength and direction of relationships, control for confounding variables, It goes beyond correlation by showing how multiple factors work together to influence behavior.

 Describe how you could use the regression equation to predict one variable from another? Be sure to explain the use of each variable in the regression equation.

to use the regression equation, you plug in a value for the predictor (X) and solve to get the predicted outcome (Y). The intercept (a) gives the baseline prediction, and the slope (b) tells you how much the outcome changes for each unit increase in the predictor.

   What is the difference between a linear regression and a multiple regression?

Linear regression predicts an outcome from one variable, while multiple regression predicts an outcome from several variables at once

 

     What are the three essential properties of an experimental design?

1. manipulation of the independent variable

2. random assignment of participants

3. control of extraneous variables (confounds)

     Provide an example of an independent variable with four different levels.

• AMOUNT OF CAFFEINE CONSUMED 

• Level 1: 0 mg (no caffeine / placebo)

• Level 2: 50 mg (about half a cup of coffee)

• Level 3: 150 mg (about a regular cup of coffee)

• Level 4: 300 mg (about two strong cups of coffee)

What is the difference between random selection and random assignment?

Random selection = how participants are chosen from the population

Random assignment = how participants are placed into groups

   Explain three different threats to internal validity. How could you reduce the impact of these?

1.confounding variables:Example: In a caffeine vs. no-caffeine memory test, if the caffeine group is always tested in the morning and the control group at night, time of day (not caffeine) could cause differences.

2.Testing Effects:Taking a pretest influences posttest performance. Participants may perform better just because they’ve seen the test before, not because of the treatment.

3.Attrition:Participants drop out of the study, and if dropout isn’t random, it can bias results.

 Describe how both confound variance (part of systematic variance) and error variance can be problematic (they have different effects, but both are bad!).

• Confound variance: Misleads you → threatens validity (you might think IV caused the effect when it didn’t).

• Error variance: Blurs the signal → threatens reliability/power (makes it harder to detect an effect)

describe the difference between a pretest-posttest design and a post-test only design. What are the strengths and weaknesses of each?

pre-test;post-test design: 

-when you are measuring the dependant variable

-Strength:looking for a change overtime

-Weakness:bc you have taken the test before, you have an advantage 

Post-test only design:

-when researchers test participants once, after the treatment, to see if the independent variable caused a difference.

-Strength:simple, time efficent, no pre test to influence behavior on test

-Weakness:you only know groups differ, not by how much,if random assignment fails, groups may not be distributed equally at start

    What is a matched subjects design?

is an experimental design where:

• Researchers pair participants together based on important traits (like age, IQ, baseline scores, or personality).

• Within each pair, one person is randomly assigned to the experimental group and the other to the control group.

  Describe the difference between an independent groups design and a repeated measures design. What are the strengths and weaknesses of each?

 

• Independent groups design = different people in each condition; avoids carryover effects but needs more participants.

• -strength:ppl are not exposed to multiple effects 

• weakness:requires more ppl

• Repeated measures design = the same people in all conditions; more powerful and efficient but risks order effects that need control.

• -strength:more statistical power and needs less ppl 

• -weakness:must use counterbalancing to control this 

   What is the “Bonferroni Adjustment”?

a way to control for Type I error (false positives) when you run multiple statistical tests at once.

What conclusion would you make if you conducted a two-tailed independent samples t-test in a study with 30 participants and ended up with a t-value of 1.92?