Investigation design, data analysis and presentation

?

Generating research aims

Starting point for psychological research study is for the researcher to think carefully about what the investigation is trying to discover, generate an appropriate aim to make the focus of the study clear.

Aim of research study may be refined into a research question, in experimental research into a hypothesis

Hypothesis:

  • a statement that is testable
  • general prediction made at the outset of an experimental investigation about what the researcher expects to happen
  • essential to phrase the hypothesis carefully so that its unambiguous and testable
  • if wording of hypothesis is too general: could be difficult to test
  • when wording is more clear, may lack general application 
1 of 11

Alternative and Null hypotheses

Alternative:

  • predicts that something other than chance played a part in producing the obtained results
  • suggests the cause for the change in the DV was the IV
  • hypotheses can either be directional or non-directional
  • Directional: predicts the direction in which results are expected to occur
  • Non-directional: does not predict the expected direction of the outcome but predicts a difference/change

Null:

  • Predicts that the results of an experiment can be explained by chance alone rather than manipulating the IV
  • null hypothesis would predict no difference between 2 conditions
  • if differences do occur, it is assumed this happens due to chance 
  • predicts alternative hypothesis is not true
  • if probability value = high, null accepted
  • if probability value = low, null rejected
2 of 11

Experimental design

Aim of successful experimental design:

  • provide overall plan for the experiment
  • ensure that appropriate and precise measurements can be made
  • enable all data collected to be analysed correctly
  • eliminate potential sources of ambiguity or bias
  • ensure high levels of control over IV and DV

3 types of experimental design:

  • Independent groups design: different participants are used in each condition of the experiment
  • Repeated measures design: same participants are used in each condition of the experiment
  • Matched pairs design: each group of participants is carefully matched on the variables considered to be relevant to the experiment
3 of 11

Independent groups design

Summary:

  • involves using different participant in each experimental condition
  • experiments using this design typically involve:
  • a control condition and one or more experimental condition or..
  • two or more experimental conditions
  • group of participants given the experimental treatment referred to as experimental group
  • group that provides comparison and no experimental treatment known as control group
  • random allocation ensures characteristics of participants do not differ systematically between conditions at the start of the study
4 of 11

Pros and Cons of Independent groups design

Advantages:

  • no problem with order effects
  • wide range of potential uses

Disadvantages:

  • potential for error resulting from individual differences between groups
  • may represent uneconomic use of participants if participants are in short supply
5 of 11

Repeated measures design

Summary:

  • involves exposing every participant to each of the experimental conditions
  • one of the conditions may be a control conidition, serves same purpose as control condition in independent groups design
  • Counterbalancing: involves equal numbers of participants undertaking required tasks in different orders
  • Randomization: involves adopting strategy for randomly determining the order of presentation of experimental conditions e.g drawing lots of tossing a coin

Advantages:

  • individual differences are removed as confounding variable

Disadvantages:

  • order effects may occur when participant takes part in more than one experimental condition and hence confound the results
6 of 11

Matched pairs design

Summary:

  • aims to achieve key advantages of independent groups and repeated measures design
  • involves matching each participant in one of the experimental condtions as closely as possible with another participant in the second condition based on all relevant variables

Advantages:

  • combined advantages of independent groups and repeated measures

Disadvantages:

  • can be difficult and time consuming
  • complete matching of participants can rarely be achieved
7 of 11

Key aspects of investigation design

Operationlising variables:

  • variable is a thing that may vary in some way and can either be categorised or measured
  • the control and manipulation of variables are central to psychological research
  • operational definitions of variables are precise descriptions of what a particular researcher understands by a particular term
  • operationalising variables often results in narrowing down the research focus
8 of 11

Conducting pilot studies

  • Important step of designing a good research study is using a pilot study
  • this is a small-scale trial run of a specific research investigation to test out the planned procedures and identify flaws and areas for improvement
  • conducted before time and money are invested in the main study
  • carried out to find out whether there are problems with:
  • design
  • clarity of instructions
  • measuring equipment
  • researcher will often ask participants taking part in the pilot study to highlight problematic areas or ambiguities during the trial run
9 of 11

Selecting participants and sampling techniques

  • General principle: the larger the sample, the more likely it is to provide an accurate estimate about the nature of the population from which it has been drawn
  • deciding sample size reflects a delicate balancing act between need to accurately represent the target population and practical considerations such as saving time and money on the other
  • samples can be selected in several different ways
  • three main methods of sampling: random sampling, opportunity sampling, volunteer sampling
  • Random: every person or item in target population has an equal chance of being selected for inclusion
  • necessary to have a list that identifies every single person/item in the target population in order to generate a random sample
  • selection must take place in an unbiased way
  • Opportunity: widely used because it's convenient. Involves researcher selecting anyone who is available to take part in a study from a given population
  • Volunteer: participants select themselves to take part in a research study often by replying to an advertisement
10 of 11

Relationship between researchers and participants

Demand characteristics: occur when participants try to make sense of a research situation they find themselves in and act accordingly

  • expected that a relationship may develop between a researcher and participants
  • research investigation is liable to be influenced by those taking part
  • this mean they act diferently to how they would outside the research situation: causes a problem
  • well designed research aims to minimize demand characteristics effects as much as possible

Investigator  effects: result from the effects of researcher's behaviour and characteristics on an investigation

  • expectation effects can occur when participant is deeply committed to achieving a particular outcome
  • naturalistic observational studies: presence of the observer can cause participants to behave in ways that are different to normal: restrained/exuberant
11 of 11

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Data analysis resources »