Writing Research Reports Flashcards
How is a research report structured
Title
Abstract
Instruction
Method
Results
Discussion
References
Appendices
Why are reports structured the way they are
It allows readers to quickly find information of interest without necessarily reading the whole article
I provides important details to allow replicability
How is a Harvard reference for a journal article structured
Author surname, initials., Author surname two, initials… Year. Title of article. Title of journal, volume (issue), page number
How is a APA journal article reference structured
Author surname, initials., Author surname two, initials… (Year). Title of article. Title of journal, volume(issue), page number
Structuring and introduction
- include why the topics interesting important
- What work has already been done in this area
- How your work builds on previous lit
- What gaps are in the knowledge and how your study fills the gap
- use a funnel shape
Population versus sample
We cannot gather data from the whole population as it is too costly and time-consuming and not all will respond.
Instead we select a sample from the population and gather data from them
Generalising from sample to population
Generalisation is the ability to apply the results from the study or experiment to the wider target population.
Random sampling.
Quite uncommon in psychology
Leads to a representative sample – confident in generalising ourselves to the entire population
Two of the most common forms of random sampling are:
- Simple random sampling
- stratified sampling (divided into meaningful groups)
Non-Random something
Very common in psychology
Leads to a less representative sample (less confident in generalising I results to the entire population), but saves time and money, and is often the only practical option when the population is large (e.g., or people living in the UK, or all males).
The three most common forms of non-London something are
- Voluntary something
- Snowball something
- Convenience something
How big should your sample be
The size of the sample determine:
- extent to which you can generalise findings
- Probability of a chance finding or missing of important finding
Depends on
- size and homogeneity of the population
- nature of the variables measured
- required precision of results
- how confident you want to be about your results.
Experiments
Are the only way to explore causal relationships
(-Do changes in variable A cause changes in variable B)
We then manipulate the values of one variable (A) and see if affects the second variable (B), keeping all other variables constant
Independent variable
The variable which you think may be a causal variable
The variable that you manipulate (for example exercise level)
Has two more conditions/groups (e.g. regular exercise versus no exercise)
Dependent variable
Dividable that you think it’s affected by the changes in their independent variable
The variable that you measure for example memory performance
Quasi experiments
the investigator has no control over the independent variable, but has power over how the dependent variable is measured.
A common example is biological sex as the independent variable
Experimental and control groups (conditions)
Experimental condition:::
(regular exercise)
The condition where the independent variable (exercise) is present
Can be several experimental groups for example once a day, once a week
Control condition::::
(no exercise)
The condition where independent variable (exercise) is absent.
Extraneous variables
Factors that can potentially affect the relationship between independent variable and the dependent variable – control for as many as possible
An extraneous variable is called a confounding variable if it differs systematically with the independent variable
Preventing extraneous/confounding variables
•Match the conditions on key variables
– balance age, sex, education ETEC
– cannot predict all possible extraneous variables
- Standardised procedure
- Randomisation of the sample to the conditions
– insures equal dispersion of all extraneous variables (even those that have not been identified)
Randomisation is the key element to true experiments!!!!!!
Between groups design
(Or independent measures/between participants/between subjects)
Compares different participants in different conditions
Advantages and disadvantages of between groups design
Advantages :::
– no carryover affects: avoids one condition contaminating other condition
– processes quicker for the participants (less likely struck out or get bored)
Disadvantages::::
– individual differences have greater affect
– need more participants
Within groups design
(Or repeated measures/between participants/between subjects)
Compare same participants in both conditions
Advantages:
- Affect of individual differences is reduced
- fewer participants needed
Disadvantages:
- boredom, fatigue, dropout
- One condition may contaminate favour – carryover affects for example practice
Counterbalancing
Helps to cancel out order affects
the participant sample is divided in half, with one half completing the two conditions in one order and the other half completing the conditions in the reverse order.
Hypothesis testing
•Null hypothesis:
- manipulation of the independent variable will have no effect on the dependent variable
- there will be no difference between the conditions
•Experimental hypothesis:
- manipulating the independent variable or cause a change in the dependent variable
- they’ll be a difference between the conditions
In predicting a difference, hypothesis or even directional (one-tailed) or nondirectional (too tired)
Operationalising your variables
It is important to operationalise your variables
This means not only been clear on what each variable is but, how to quantify the variable you are measuring (the DV)