research methods

what are the four types of experiment

• laboratory (controlled environment)

• field (real world setting with manipulated IV)

• natural (where the IV is a naturally occurring event)

• quasi (uses pre existing characteristics)

identification of operationalised IV and DV

describe them with precise details on how the IV is manipulated and how the DV is measured

how to write different hypothesis

• think of your specific research question

• identify your variables

• state your prediction

• choose between a directional and non directional hypothesis

when do you use a directional/non directional hypotheiss

• you use directional hypothesis when you want to predict the specific outcome (one tailed test)

• you use non directional hypothesis when you want to predict the different but not the direction (two tailed test)

what are the three types of experimental design and how do they work

• matched participant ( pairs of participants are matched in terms of variables for example age)

• independent groups ( different participants are used in each condition of the IV)

• repeated measures (same participants are used in each IV condition)

give pro and con of each experimental design

• independent groups

-pro is it avoids order effects as people participate in one condition only

-con is more people are needed so it’s time consuming

• repeated measures

-pro is participant variables are reduced

-con is there may be order effects

• matched pairs

-pro is reduced participant variables

-con is very time consuming

what is counterbalancing

where the sample is split into two groups to eliminate order effects , although order effects occur for each participant they balance each other out in the results because they occur equally in both groups

what are extraneous and confounding variables , give examples and how to control them

• extraneous variable is a variable that can influence the DV in a study (age,gender)

-to control you use standardised procedures , random assigning and counterbalancing

• confounding variable is a variable that creates a misleading relationship between the IV AND DV (job or years of experience in a study of education and income)

-you can control through design methods like randomisation and matched pairs

what are the 5 different sampling techniques and how do they work

• random sampling (every member has equal chance of selection)

• systematic (where every nth number is chosen from a list)

• stratified (population is divided into subgroups and samples from each)

• volunteer (participants choose to be in study after seeing advertisement)

• opportunity (where researcher selects participants based on their availability)

give a pro and con of each sampling technique

• random - pro is no bias , con is can be time consuming

• stratifided - pro is its representative , con is time consuming

• systematic - pro is cost effective , con is risk of bias

• volunteer - pro is less experimenter bias, con is not representative

• opportunity - pro is cost effective , con is researcher bias

what are the ethical guidelines and why are they important

• informed consent

• protection from harm

• right to withdraw

• confidentiality

• deception

• debriefing

theyre imporant so it makes sure participants are protected and to make sure research is credible and reliable

who governs the ethical guidelines in the UK

the british psychological society

how does a psychologist ensure adherence

• submitting a research proposal

• detail ethical procedures

• gain approval

• adapt if needed

what is a pilot study and why do we use them?

• a small scale study to test whether critical components of the main study will be feasible and to identify any potential problems so they can be rectified before full study

• to check if the instructions are easy to understand and followed correctly by all participants and to check if the task is appropriate

what are the 3 measures of central tendency and how do you calculate them?

• mean (add all numbers then divide by the amount of numbers there are)

• median (calculate what the middle number is)

• mode (identify the numbers that appear the most)

what are the strengths and limitations of each?

• mean - strength is it takes account of all scores , limitation is very small or very large values can affect mean

• median - strength is it’s not affected by very large or small values , limitation is it doesn’t take all scores into account

• mode - strength is it’s easy to calculate , limitation is it doesn’t take all scores into account

what are the 2 measures of dispersion and when do you use each one?

• range (when the level of data is ordinal)

• standard deviation ( when the level of data is interval)

how do you calculate the range?

highest value take away the lowest value

what is standard deviation?

a sophisticated calculation that establishes how far on average the scores deviate or vary from the mean

how to correctly draw and label bar graph

• shows continuous data so bars should touch each other

• X axis is made up of equal sized intervals

• Y axis refers to frequency

how to correctly draw and label histogram

• used when data is divided into categories

• bars will always be separate in a bar graph to show you’re dealing with separate categories that are discrete

how to correctly draw and label line graph

• used to display how things change over time

• IV is plotted on x axis

• DV is plotted on y axis

why do we use inferential statistics?

to determine whether the null hypothesis can be rejected or if it must be accepted according to probability

how do you calculate the value of S in a sign test

from the table add up the number of pluses and minuses , the less frequent sign becomes the calculated value of S

how do you figure out critical values

use the table and using the number N ( number of participants used) and level of significance and if the test was one tailed or two tailed see what your critical value is

how do you compare the critical and calculated value using the word gReateR?

if the test has the letter R in the title the calculated value needs to be equal to or greater than the critical value for the results to be significant and the null hypothesis to be rejected

what does the outcome mean in terms of significance/null hypothesis plus type 1 and type 2 errors?

• by using greater if the calculated value is greater/ less or equal to the critical value the results will be significant and the null hypothesis will be rejected

• type 1 error is false positive when the researcher rejects the null hypothesis when it should have been accepted (0.1 has been used)

• type 2 error is false negative when the researcher accepts the null hypothesis when it should have been rejected (0.01 has been applied)

what is a correlation analysis and how does it differ from an experiment?

• established the relationship between two variables

• in an experiment the researcher controls or identifies the IV in order to measure the effect on the DV which allows researcher to establish cause and effect however correlations do not include an IV or DV you cannot infer cause cause and effect you just simply find a relationship between two variables

how do you draw and label a scattergraph

• label axis

• include title of graph

• correct plotting

• correct layout and use of space

what does the correlation coefficient tell you about the direction and strength of a correlation

• it leads to a number between +1 to -1 , towards the -1 end it’s perfect negative and towards the +1 end it’s perfect positive

• you need to ignore the minus sign when comparing it to a critical value if it’s a negative correlation

what are the 4 types of observation

• controlled observation

• naturalistic observation

• participant observation

• non participant observation

• overt observation

• covert observation

what do structured observations require

• behaviour checklists which include behaviour categories (operationalised behaviours you’re looking for)

• time sampling (time frames)

• event sampling (tallying off when you see the behaviour)

how do you ensure internal validity through face validity checks and pilot studies

• ask an expert to check if the behaviour categories look correct

• training observers so they are familiar with behaviour categories

• pilot study - checking whether the behaviours are correct and operationalised effectively

what is inter observer reliability and how does it work

• assessing the extent to which there is agreement between two or more observers

• results fron the behaviour checklist should be compared

what do you use to compare and what score do you need to get above to conclude the results are reliable

• a correlation test such as spearman’s rho

• if the results show .8 or above this indicated 80% agreement and demonstrates observers are consistent

how to choose correct test

cats sing classical music while sitting under rainy porch