Lesson 2 Internal and External Validity PDF

Let’s share some of our Experiments Laboratory Field Experiments Experiments In Participants Natural Artifical Setting Setting High Levels of Also manipulate Standardisation Indepen...

Let’s share some of our Experiments Laboratory Field Experiments Experiments In Participants Natural Artifical Setting Setting High Levels of Also manipulate Standardisation Independent Variable Controls Also Measure Dependent Variable Have Independent Variable An attempt to Control Have Depdendent Variable Extraneous variables – But difficult More room for Situational Question: Which is better? Variables to effect DV What are the problems with laboratory experiments?  Laboratory Experiments are NOT natural Problems with The more you control extraneous variables the less realistic the Laborator experiment comes y Controlling extraneous variables means we can be sure our IV Experime manipulation is causing the change in DV nts But! this means the situation (experiment) is unnatural and therefore doesn’t generalise to real life What are the problems with Field experiments ?  Can’t control all variables affecting the DV Problems The changes in the DV could be caused by one of many situational with Field variables, real life is chaotic Experime Lose control of clear Cause & Effect nts Cannot necessarily conclude that our measured effect (DV) happened because of our manipulated variable (IV)  A New Concept: Validity If a study is valid then it truly represents what it was intended to represent. Experimental validity refers to the manner in which variables that influence both the results of the research and the generalizability to the population at large. What do we think this means? Validity refers to how well a test actually measures what it was created to measure. What do we think this means? Lesson 2: Internal Validity: The extent to which a study establishes a Internal trustworthy cause-and-effect relationship between a Validity treatment and an outcome & External Validity: How well the experiment can be applied External to other settings Validity Ecological Validity: The extent to which we can generalise the findings to real life (specifically) Let’s ask a research question: Do violent video games make people more violent? So… What do we think about this experiment? What can we conclude? Evaluation on the Experiment We will definitely be able to measure the effect of video games on aggression However, the situation is so unnatural it may not represent reality In the real world, there are variety of factors affecting how we react to stimuli These could cancel out the effect (In other situations/real life) We can conclude violent video games can make you more aggressive but we cannot conclude that this would necessarily happen in real life The study has HIGH Internal Validity, LOW External (AND Ecological) Validity  We have a Are we sure we are spectrum measuring what we want to Does the experiment show natural human behaviour in a measure? natural setting High High External / Internal Ecological Validity Validity Laboratory Field Experiment Natural Experiment Experiment (No This is a rule of thumb, there are a thousand exceptions to all of this. E.g. longer on Natural experiments can lack generalisability/ecological validity because they are syllabus) too specific or unique. For example we may want to study peoples experiences of working from home during the coronavirus. The problem is the coronavirus is an unusual setting for working from home to happen in. we wouldn’t be able to tell if Psychologists need to use a variety of methods to answer their research questions. This is called Triangulation.  Activity Time: 1. Look at the attached document, with examples of studies, read through them (you don’t need to take notes) and decide whether they are natural/lab or field experiments 2. Finally choose a new research topic, design three different experiments that would examine that topic (one lab, one natural, one field) however this time you will create 3 experiments for it: IV/DV (for the ones that need it Study type – interviews/experiments etc Method/materials/etc Research design (independent measures, etc) – if needed How will you control/not control extraneous variables Homework – Read through Page 10 - Validity This reading expands on what we have done today. Make notes on: Subjectivity Objectivity The types of Valdity Demand Characteristics Generalisability Make sure you have notes on – Strengths and Weaknesses of Lab & Field Experiments Comparison of Lab and Field Experiments