Turn on thread page Beta

What is an experimental design? watch

Announcements
    • Thread Starter
    Offline

    8
    ReputationRep:
    Hey, I'm Paige!

    I'm currently doing AS psychology, even though i'm restarting my a levels this year as I'm moving college, but theres one thing that keeps popping up that I really do not understand...

    What is an experimental design?
    What types are there?
    and what are advantages and disadvantages of them?

    Due to personal reasons I was away when my college did the research methods topic, so i've missed a lot.

    If anyone could help me it'd be much appreciated.
    On here or mail me.
    Thanks!
    x
    Offline

    0
    ReputationRep:
    Briefly, an experimental design is precisely that. One that uses an experiment. That is to say, a direct manipulation of one variable to evaluate any differences produced in another variable(s).

    Advantages being you can (hopefully) infer directional causation instead of just correlation. Disadvantages come in the fact that some variables of interest are not manipulatable (or ethical to do so) making it difficult to assess causation. Additionally, lab experiments can be quite contrived and detached from reality making extrapolation difficult.

    There are many many types of experimental design. Better off referring to a text for descriptions of the main themes.
    Offline

    0
    ReputationRep:
    (Original post by PaigeSnow)
    Hey, I'm Paige!

    I'm currently doing AS psychology, even though i'm restarting my a levels this year as I'm moving college, but theres one thing that keeps popping up that I really do not understand...

    What is an experimental design?
    What types are there?
    and what are advantages and disadvantages of them?

    Due to personal reasons I was away when my college did the research methods topic, so i've missed a lot.

    If anyone could help me it'd be much appreciated.
    On here or mail me.
    Thanks!
    x
    Generally...
    Independent group design - participants take part in only one condition
    Repeated measures design - participants takes part in both conditions
    Matched pairs design - a participant is matched with another participant on several factors (e.g. age, gender) and they are put in different conditions

    As for advantages/disadvantages, check a textbook. Most are pretty straight forward though e.g. a good thing about repeated measures design is that you need less participants
    • Thread Starter
    Offline

    8
    ReputationRep:
    Thank you Phalange!
    One more thing though, what do you mean by ' in only one CONDITION' , 'in both CONDITIONS'
    do you mean like they take part in one experiment or 2 experiments?
    I probably sound proper thick, but it's better to clear it up than to do a guessing game.
    :L
    x
    • PS Helper
    Offline

    2
    ReputationRep:
    PS Helper
    (Original post by PaigeSnow)
    Thank you Phalange!
    One more thing though, what do you mean by ' in only one CONDITION' , 'in both CONDITIONS'
    do you mean like they take part in one experiment or 2 experiments?
    I probably sound proper thick, but it's better to clear it up than to do a guessing game.
    :L
    x
    In an experimental design, there are usually two main ways in which participants are arranged into each condition of your experiment.

    Independant-measures/ Between-subjects - This is where each participant takes part in only one condition.

    Repeated-measures/ Within-subjects - This is where a participant will take part in every condition of your experiment.

    There are situations where one design would have an advantage over the other.

    I'll use an example: You want to measure the effects of sleep deprivation on cognitive functioning. Something simple like a memory test where you are given twenty pairs of numbers to recall once they have all been presented. Lets say we have six participants - John, Mike, Julie, Wendy, Will and Mary.

    You have three levels of your IV - Control (no sleep deprivation - so 8 hours sleep), 4 hours sleep, and 0 hours sleep. Your DV will be the amount of correct pairs recalled.

    With an independant-measures design, John and Mike are put into the control group, Julie and Wendy are put into the 4 hours sleep group, Will and Mary are put into the 0 hours sleep group.

    I'll use another example: You want to work out whether a new anti-depressant has an effect on the participant. Your IV is time (measured in weeks with 3 levels - control, 2 weeks and 4 weeks) and your DV could be something like the number of "sad" thoughts experienced by the participant.

    In this example, all six of your participants will be measured across all three of your time periods.

    The "condition" refers to the group in which the participant has been placed.
    Offline

    0
    ReputationRep:
    (Original post by PaigeSnow)
    Thank you Phalange!
    One more thing though, what do you mean by ' in only one CONDITION' , 'in both CONDITIONS'
    do you mean like they take part in one experiment or 2 experiments?
    I probably sound proper thick, but it's better to clear it up than to do a guessing game.
    :L
    x
    Dillio summed it up. If you are still confused...

    I want to do an experiment to see if people who listen to rock music do better at a memory game than people who don't listen to any music.

    First condition: Participants listen to rock music whilst words appear on screen. They are then asked to write on a piece of paper any words they remember

    Second condition: Same as above except participants do not listen to ANY music whilst doing the task.

    My independent variable (the thing I manipulate) is whether participants listen to rock music or no music.
    My dependent variable (the thing I measure) is the amount of correctly recalled words

    -----
    I have 10 participants.
    If I used an independent group design, 5 participants will be in the first condition and the other 5 will be in the second condition.

    If I used a repeated measures design, all 10 participants will be used in both conditions.

    If I used a matched pairs design, I would get the 10 participants and then match them to 10 OTHER participants based on stuff like age, gender. So I have a pair.
    For example, 2 of my participants are:
    1) John, aged 40
    2) Aysha, aged 20

    For matched pairs, I will find two other people who are similar to John and Aysha. John and Aysha will be put in the first condition whilst the 2 other people I matched will be put in the second condition.
    • Thread Starter
    Offline

    8
    ReputationRep:
    Oooh, thank you so much! :')
    No confusion any more for me haha. :')
    I really do appreciate it, most people on here would just take the piss, from what i've seen so far anyway :L
    Thanks!!!
    Paige x
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: April 17, 2011

University open days

  • University of Warwick
    Undergraduate Open Days Undergraduate
    Sat, 20 Oct '18
  • University of Sheffield
    Undergraduate Open Days Undergraduate
    Sat, 20 Oct '18
  • Edge Hill University
    Faculty of Health and Social Care Undergraduate
    Sat, 20 Oct '18
Poll
Who is most responsible for your success at university

The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

Write a reply...
Reply
Hide
Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.