Research methods 2 Flashcards
what is qualitative data?
data expressed in words
may be written description of thoughts, feelings etc
what is quantitative data?
data expressed numerically
what is secondary data?
data collected by someone other than the researcher carrying out the current research
what is primary data?
original data collected specificallty for the purose of the research
what are the steps involved in using a sign test to measure significance?
1.work out diff between the two conditions ( e.g condition A - CONDITION B )
2.if the scores is positive put + at end if negative -, if no diff cross put
3.count the no. of + and - the number that is SMALLER = calc value (s) of test
4.go to critcal values table ( should know calc value, tail hypothesis, level of significane and no. of p’s
TAKE AWAY THE NO DIFF FP’S FROM NO. OF P’S
5. calc value must be EQUAL TO OR LESS THAN THE critical value for signifcance
how is a significant diff answer written?
there is no significance this is because the cal value of .. is not eqaul to or less than the critical value of … at 0.05 level of significance for a ( type of test)
therefore we should accept the null hypothesis and reject the alternative
analyse qualitative data
what is a content analysis
identify coding categories
- converting qualitative data into quantitative data through recording the frequency of coding categories
1st researcher reads through the material
2nd identifies reccuring coding categories
3rd reads again- tallies how often these categories occur
this allows researcher to compare
analyse qualitative data
what is coding in a content analysis?
established by researcher after looking through the qualitative data
* simply counting how many times and word/phrase appears
analyse qualitative data
what is a thematic analysis?
identify themes and see how they relate to eachother
- data remains qualitative
- once data is transcribed it is reviewed repeatedly so that researcher can identify trends in the meaning conveyed by language
- identify particular themes,see how they relate to eachother and how they may influence the behaviour of intrest
what are case studies
in-depth study or investigation of a single person, group,event or community
qulitative research method
* data is gathered from a variety of sources using diff methods over a period of time maybe so processes and developments can be studied as they happen
what are two examples of case studies?
HM- underwent surgery as treatment for epilepsy part of his hippocampi removed= unable to form new long term memory but stm intact= support for msm of memory as it shows that the memory stores are seperate and independent
**little hans- ** a five year old boy who had a phobia of horses displaced his fear of his father onto horses (symbolic representation of his real fear being castrated by father) support for the oedipus complex
how findings are reported
what is the format of scientific reports in order ?
- abstract
- introduction
- methods
- results
- discussion
- references
how findings are reported
what are the things featured in the method sections
particpants
design
materials
procedure
ethics
why does it matter if research is accurate
dogy research from e.g andrew wakefield can have huge implications psychology needs a way of protecting itself and maintaining intergrity
what is peer review
a process that takes place before a study is published after research is conducted to ensure that the research is of high quality, contributes to the field of research and accurately presented
what is the purpose of peer reviews
• protect the integrity and credibility of psychology
• ensure only accurate scientific information is shared
• assess the originality of research
what is the process of peer reviews
• another researcher in the same field reviews the research
• check the method is valid and free from bias
• ensure appropriate data analysis and free from bias
• assess originality
•peer recommends its published, not published or can be re-submitted after revisions
• editor decides if its published
what are research proposals use
purpose- to evaluate proposed desgins for reserach funding
process- submitted to panel of peers, specialise in field, and assessed for merit
what are economical implications of research
positive or negative
e.g treatments for mh conditions- research reduces absences, early interference etc positive for the economy
features of science
what are the features of science?
- theory construction
- hypothesis testing
- Empirical evidence
- Paradigm (paradigm shifts Kuhn)
- Replicability (reliability)
- Objectivity
- Falsifiability (Popper)
design your own study
what should be involved in a design your own study question?
data analysis- descriptive (mean median mode, range standard deviation) and infrential analysis (statistical test)
* follow all bullet points
* how process and why ( justification)
what is a meta-analysis?
uses secondary data
* researcher identifies studies which have investigated the same aims/hypotheses
* all studies are combine them all together for analysis and draw an overall conclusion
what are descriptive stats and what are the two examples and those examples?
analyse data to help decribe,shows,summerise it in a meaningful way
1. measures of central tendency- mean meadian mode
2. measures of dispersion- range and standard deviation
central tendency
how is the mean calculated?
adding up all the scores and dividing it by the number of data points
can only be used when the data is interval level
central tendency
what is the median and how is it calculated?
middle score- put all in numerical order and cross of until get to middle
if there are two numbers in middle, use the half way pont between the two e.g 9 and 10 median is 9.5 OR add together divide by 2
ordinal level
central tendency
what is the mode?
most common number
nominal data
sometimes two = bimodal
what do measures of dispersion do
describe the spread of data around a central value (mean mode median)
tell us how much variety there is in data the data
measures of dispersion
what is the range?
the difference between the highest and lowest values
ordinal levels
measures of dispersion
what is standard deviation?
the spread of scores around the mean (indication of how far the average particpat deviates from the mean)
* Small SD- the most scores cluster around the mean score, little variation between scores and mean- suggests the effect of iv on dv is consisitent
* large SD- there is greater vairety in the scores and the mean is not representitive- suggets the effect of the iv on dv is inconsistent
how to answer a question regarding mean?
the iv has a ( positive or negative impact on the dv ….. this is shown by the mean score of condition …. being … higher or lower than condition …
how to answer a question regarding standard deviation
?
the effect of iv on the dv is inconsistent or consistent this is shown by condition …. having a sd of …. and the condition having an sd of ….
data presented in a graph
what are the basics for graphs?
- axes must be labelled ( x horizontal, y verical)
- appropriate keys
- title with ref to iv and dv
- appropriate scale e.g 8, 10,12 etc not 8 11 ,15 20
data presented in a graph
what do scattergrams show?
- correlation between 2 variables
- a dot or cross
- ONLY USED WITH CORRELATIONS
- both axis have numbers
data presented in a graph
why are bar charts used?
- when the data can be divided into categories- go on the x axis and numericals on y axisis
- keep seperate- to show seperate conditions
- only put together if there are 2 ivs or dvs- one category on bottom and other represetned by colour
- nominal data
data presented in a graph
when are histograms used
continuous equal intervals
- for continuous data
- x axisis- continous intervals should be equal size e.g 0-9, 10-19 etc
- y axis- frequency within each interval
- bars must touch
data presented in a graph
when are line graphs used
- when data is continuous but not based on intervals
- points connected with a single line to show how something changes over time
data presented in a graph
when are pie charts used
discrete data- categories
•each slice represents a proprtion of a whole
•first work out frequency by adding all scores up
•then work out what % of total frequency is made up by each category e.g 6/60x100
what are normal distributions
a systemetrical bell shaped curve in which most people occupy the middle area
* the mean median and mode are equally the same
what is a positive skew
mean is higher than the mode and median- dragged up by few higher scores
most people scores are low
what is a negative skew
mean is lower than the mode and median- dragged down by the few lower scores
most people get high scores
how is significance determined
using infrential stats- statistical tests
what three things need to be considered to decide what stats test to use?
- level of data- diff types of quantitaive data they are the dv/covariables
- difference or correlation
- related or unrealted designs
stats test
what is nominal data?
category based and focuses on the frequency of each category
stats test
what is ordinal data
data placed in rank order on a scale OR any subjective scale e.g hunger etc based on personal interpritation
stats test
what is interval data?
data placed on a scale where the intervals are fixed e.g height or time in secs
these are universal accepted forms of measurements - non arbitary- cannot be affected by personal interpritation
stats test
what is a difference and a correlation
effect of iv on dv or difference between one thing and another thing
relationship between two variables
stats test
what is the difference between unrealted and related samples?
related- praticpants are the same e.g matched pairs or repeated measures design
matched pairs because they have been matched on a key variable= related
unrealted- particpants are different in each condition e.g independent groups design
significance
why are inferntial stats used as?
a way of knowing if the diff between two conditions is due to the iv affecting the dv or due to chance
carrying them out= see if we can be 95% sure results are not due chance
significance
what is probability
the likelihood
significance
what is a null hypothesis and when should it be rejected and accepted
there will be no significance/difference
reject when there is a difference/significance
accept when there is no differrnce/significance
significance
what is an alternative hypothesis and when should it be rejected and accepted
predicts there will be a difference/significance
reject when there is no diff/signific
accept when there is a diff/signific
significance
what is p
the likelihood that the results are due to chance rather than the iv affecting the dv
so p
significance
how sure do we want to be that we are not making errors
95% sure that findings not due to chance
5% probalility that it is due to chance
significance
what value does p always need to be
lower than 0.05
but when humans are involved 0.001= 99.9% sure not due to chance
significance
what is the probalility level that psychs always use
p≤0.05
* the probalitity that we are in error when we say that the diff is significant is 5% or less
significance
what are the 3 picies of info needed to read a critical value table?
- hypothesis- one tailed or two tailed
- how many p’s are there ( N or Df)
- level of probalitity used ALWAYS USE 0.05 unless told otherwise
significance
what are the steps for a critical value table?
- identify the hypothesis
- identify calc value
- identify level of significance
- use table to find out critical value
- determine wheter it is significant or non-signifcant
significance
what are type 1 errors?
there was a significant difference when actually there wasnt one
accepting alternative hypothesis but should have been rejected
rejecting null but should have been accepted
p=0.05= 5% probability making type 1 error
significance
what is a type 2 error
there was no significant difference when actually there was one
accepting null hypothesis but should have been rejected
rejecting alternative- should have accepted it
significance
what are the consequences of type I errors
- publishing of false data
significance
what are the consequences of type II errors
- researcher may think that their study was not good enough to find an effect
- researcher may give up on reserach that may actualy provide useful important findings
significance
how to avoid making a type I and II error
- I- using stringent significance level p= 0.01= 1% chance of making type 1 error
- II- using a more leninet significnace level p= 0.1
- both- increase the chance if other type error being made so 0.05 balances out these risks and is used unless socially sensitive research
how findings are reported
what do the first three parts of a scientific report do?
- brief summary of study inclusing aims, methods, results and conclusions
- discusses previous reserch, includes theories and previous studies ( identify issues with them)- uses existing research to show how reached hypothesis
- participants, desgin, materials, procedure, ethics
how findings are reported
what do the last the last three parts do?
- results of the study- central tendancy and dsipersion ( descriptive stats), graphs, inferential stats
- summary of findings, discuss implications, practical issues, strangths and limitations of methods, future research that may follow on
- authour date titile location publisher
what is included in the reference section of a scientific report?
animals
dont
talk
like
people
