Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
82 Cards in this Set
- Front
- Back
Qualitative Research |
An umbrella term which the primary goal is to understand and explain social phenomena. Based on non-numerical data obtained in a natural setting through extensive observations and interviews |
|
What are the 5 phases of qualitative research? |
1) The Researcher as a multicultural subject: The researchers are not separate from the research process. They are multicultural subjects who are part of the data collection process. As such, researchers must be aware of and acknowledge their multiple identities and their experiences of the phenomenon. 2) Theoretical Perspectives and Paradigms: Positivist vs Constructivist. Quantitative research follows a positivist line of thought. Uses inductive (specific to general) processes to uncover concepts and build theory. 3) Research Tradition: 5 different ones: Basic, Narrative, phenomenological, grounded theory, and case study. 4) Methods of Collection and Analysis: 4 things to consider: locating and accessing the data site, determining sample technique (purposive sampling usually), collecting and recording data, and maintaining ethics and confidentiality. 5) Interpretation and Evaluation: Verification of data (prolonged engagement, triangulation, peer review, and negative case), interpreting themes, further knowledge and build theory by verifying data. |
|
Researchers as a multicultural subject |
Qualitative research Researchers are not separate from the research process. They are multicultural subjects who are part of the data collection process. Therefore, researchers must be aware of and acknowledge their multiple identities and their experiences of the phenomenon. |
|
Positivist |
Argues that knowledge is derived from logical and mathematical thinking as well as sensory experience. Very objective. Believe the world and society operate according to physical laws. Quantitative research is based around this idea |
|
Construtivist |
Argues that knowledge is generated through the interaction of human experience (environment) and a person's ideas (process of thought). The mind gives meaning to reality, ie. Reality is socially constructed. Thus, multiple realities exist because reality is based on historical, social, and temporal factors of each individual. This describes Qualitative research (subjective, non-numerical, non-statistical, small samples, open-ended, narrative for results.). Focuses on meaning and understanding through rich description. Uses inductive (specific to general) reasoning to uncover concepts |
|
Inductive Reasoning |
specific to general. Scientific method based on this. |
|
What are the 5 research traditions of qualitative research? |
1) Basic 2) Narrative 3) Phenomenological 4) Grounded theory 5) Case study |
|
Basic Qualitative Research |
Uses interpretive paradigms to descriptively explore a phenomenon |
|
Narrative Qualitative Research |
Researchers collect descriptions of events to create stories from those descriptions with a plot line that has a beginning, middle, and end. |
|
Phenomenological Qualitative Research |
Researchers are interesting in common and shared experiences of people in order to determine the 'essence' or core of human experience. Uses in-depth interviews and observations |
|
Grounded Theory Qualitative Research |
Researchers generate theory from views of participants in the study. Analysis is systematic comparitive and the research is "grounded" in real-world experience. |
|
Case-Study Qualitative Research |
An in-depth exploration of a single case, program, or event. Typically seen in medicine |
|
What are 4 things to consider when looking at methods of collection and analysis in qualitative research? |
1) Locating and accessing the data site 2) Determining the sample technique 3) Collecting and recording the data 4) Maintaining eithics and confidentiality |
|
What are things to consider in qualitative research when locating and accessing a data site? |
- Are the demographics (age, sex, race) suitable for the research question(s)? - The researcher may be more successful if they possess similar characteristics to the participants in question (ie Native American) - To help with the above point, the researcher can find a gate keeper to help them access the data site. |
|
What are things to consider in qualitative research when determining the sampling technique? |
After accessing the data site: -Purposive sampling is the most common -Sampling is sufficient when no new themes or answers emerge |
|
What are things to consider in qualitative research when collecting and recording the data? |
The most used collection and recording techniques are photographs, videos, letters... Observation is also used to see social interactions, patterns, conversations, or events... -Interviews (semi-structured) are the most common method when the behavior cannot be observed |
|
What are the keys to a successful interview? |
Questions should be -short and concise -open-ended -no leading questions -minimize the use of 'why' at the beginning of questions -Avoid 2 questions in one |
|
How do you engage in an interview? |
- Provide the purpose and procedures (letter of information - Schedule a convenient time for the participant - Give informed consent to be signed - Bring questionnaires, tape recorder, interview, guide etc. (be prepared) - Be punctual -Build and maintain rapport -Maintain neutrality -At the end of the interview be bried and thank the participants. |
|
What are things to consider in qualitative research when maintaining ethics and confidentiality? |
- Since rapport was established, confidentiality is crucial. - Use pseudonyms for participants; do not use their names -Results from the interviews affect the researchers since the researchers are the research instrument -Remain ethical when pushing for information, particularly if material being addressed is sensitive. |
|
What are things to consider in qualitative research when analyzing data? |
- Data Reading/Memo writing: Must read and reread the transcripts and then write memos to understad the information from the interviews. - Data Reduction: Describes, indexes, simplifies, and transforms raw data - often using a coding technique. - Coding: Organizes the data by assigning shorthand designations. Each segment or code is a word or short phrase that suggest how data segments inform the objectives or questions. - Next step is to relate the codes to each other and then find the higher order themes. |
|
What is the CREATIVE approach in qualitative research? |
Consider the research question Read literature Examine data for info relevant to question Assign label to units to capture meaning Thematize data (make themes) Interpret emerging themes (relevant to Q) Verify data Engage in writing of information found |
|
What are the ways to verify qualitative data? |
1) Prolonged Engagement: Sufficient time the research spends immersed in the data site, culture, and saturation of the data. 2) Triangulation: Using multiple methods to collect data (ie interview, observation, and documents) 3) Peer Review: External check with a colleague who's familiar with the research. 4) Negative Case Analysis: Researchers explore cases that don't fit the analysis |
|
Verification |
Confirm the trustworthiness of data |
|
Prolonged Engagement |
A method of qualitative verification. Sufficient time the research spends immersed in the data site, culture, and saturation of the data. |
|
Triangulation |
A method of qualitative verification. Using multiple methods to collect data (ie interview, observation, and documents) |
|
Peer Review |
A method of qualitative verification. External check with a colleague who's familiar with the research. |
|
Negative Case Analysis |
A method of qualitative verification. Researchers explore cases that don't fit the analysis |
|
Why is data analysis important? |
Simply listing scores is meaningless. It is necessary for comparison between groups or to describe the performance of an individual within a group. |
|
Statistics |
Values calculated using information obtained from the sample which is used to estimate information about the population (parameters) 2 types: Descriptive and inferential |
|
Population Parameters |
Information about the population from statistical data collected |
|
Descriptive Statistics |
Describes characteristics of group |
|
Inferential Statistics |
Inferences from a sample to a population |
|
How can scores be classified? (Scales of Measurement) |
Nominal Ordinal Interval Ratio |
|
Nominal scale of measurement |
Classifies objects in accordance to similarities and differences with respect to some property. Scores cannot be hierarchically arranged and are mutually exclusive ie sex, race... |
|
Ordinal scale of measurement |
Scores can be organized hierarchically but do NOT have a common unit of measurement between each score. ie finishing place in a race (1st, 2nd, 3rd... but the time between each isn't the same) |
|
Interval scale of measurement |
These scores have a common unit of measurement between each score, but they do NOT have a true zero point ie time, temperature... |
|
Ratio scale of measurement |
These scores have a common unit of measurement between each score and do have a true zero point. This allows comparisons such as being twice as high or 1/2 as much. ie reaction time, ruler, GPA. |
|
What is a simple frequency distribution? |
A ordered list of scores and their frequencies of appearance in the data set. Can extract the range, frequency, number of people... |
|
What is a group frequency distribution? |
This allows for a frequency visualization according to bins. Connecting the plotted frequency for the scores produces a graph known as a frequency polygon or a histogram. Smoothing this curve tells the nature of the distribution (creating a normal curve) |
|
frequency polygon |
|
|
histogram |
|
|
Normal Curve |
A group frequency distribution Bell shaped. Symmetrically centered. Defined base to height ratio. |
|
Skewed Curve |
When the graph of data do NOT resemble the normal curve. 2 types: Positive and negative |
|
Negatively skewed curve |
When the smoothed graph has a long low tail on the left indicating few participants recieved low scores (more on the positive x) |
|
Positively skewed curve |
When the tail of the curve is on the right (few participants recieved high scores). |
|
Mesokurtic curve |
Normal bell shaped |
|
Platykurtic curve |
less sharply curved (flattened) |
|
Leptokurtic curve |
More sharply peaked curve |
|
What are descriptive values? |
They summarize set of scores collected and give it meaning. They will describe performance of a group and how it compares to another. 2 types: Measures of central tendency and measures of variability |
|
What are measures of central tendency? |
Indicate the points at which scores tend to be concentrated. ie mode, media, mean |
|
What are measures of variability? |
These indicate the data in terms of their spread or herterogeneity |
|
Mode |
measure of central tendency Score that is the most frequently received in the sample. Can be used with nominal data. There can be more than one mode. |
|
Median |
measure of central tendency Middle score. Half of the scores fall above the median and half below. Data must at least be ordinal (cannot use nominal). Scores must be ordered hierarchically. |
|
Mean (x-bar) |
measure of central tendency The sum of scores divided by the numer of scores. Appropriate for interval and ratio data. Affected by value of each score. This is the most appropriate measure of central tendency |
|
If a curve is positively skewed, does this make the median or the mean larger? |
The mean is greater than the median |
|
If a curve is negatively skewed, does this make the median or the mean larger? |
The mean is less than than the median |
|
Range |
Measure of Variability The difference between the highest and the lowest score |
|
Standard deviation |
Measure of Variability The amount that all scores differ from the mean. The more the scores differ from the mean, the larger the stdev. s = sqrt( (sum(x-xbar)^2)/(n-1)) |
|
Data outliers |
A score that is more than 3 standard deviations from the mean. Can be removed from data sets |
|
What % of data falls within 1 SD of the mean? |
~68% |
|
What % of data falls within 2 SD of the mean? |
~95% |
|
What % of data falls within 3 SD of the mean? |
99% |
|
How do you measure the position of one data point in a group? |
Percentile ranks/percentiles - A descriptive value that indicates the % of participants below a designated score |
|
What are standard scores? |
They 'standardize' scores from multiple tests to a common unit of measurement. z-scores are typically used: z = (x-xbar)/SD A z-score indicates how many SD a test score is from the mean. Can easliy see outliers with this |
|
How can you find a relationship among variables? |
2 main techniques: Graphing correlation |
|
Graphing technique to find relationships between variables |
The graph shows a relationship between two variable. Based on coordinate system in which the values of a variable are listed along the x axis and the values of the other are on the y-axis ie scatter plot/gram - Can us the line of best fit, straight line, or regression line |
|
Positive Relationship |
When large scores on one measure are associated with large scores on the other measure |
|
Negative Relationship |
When large scores of one measure are associated with small scores on the other measure. |
|
Correlation to determine relationships |
Family of statistical techniques that are used to determine the degree of the relationship between 2 or more variables Things to consider are: - Correlation coefficient: (r) Can range from -1 to 1. Has 2 characteristics: Direction and strength of relationship -Direction: Whether correlation coefficient is negative of positive - Strength: How close r is to 1 or -1 (r=1 is a perfect correlation) r = (Sum((x-xbar)(y-ybar)))/(sqrt((Sum(x-xbar)^2)(Sum(y-ybar)^2)) Correlation DOES NOT imply cause and effect relationships |
|
Does correlation imply cause and effect? |
NO!!!! |
|
Correlation coefficient |
(r) Can range from -1 to 1. Has 2 characteristics: Direction and strength of relationship -Direction: Whether correlation coefficient is negative of positive - Strength: How close r is to 1 or -1 (r=1 is a perfect correlation) |
|
Spurious Correlations |
Related variables that have a high r but there is not real correlation |
|
Coefficient of Determination |
The amount of variability in one measure that's explained by the other measure. It is the square of the correlation coefficient ie if r^2 = 0.62, 62% of the variability in the x-scores is due to individuals having different y-scores. ie. x and y share 62% ov variance |
|
What factors effect the correlation coefficient? |
Sample size: r^2 with more subjects (often hundreds) Reliability of Scores: Low reliability reduces r^2 Range of scores: r^2 is smaller for homogenous groups |
|
Sampling Error |
Difference among group means because the samples are not representative of the population. |
|
What is the hypothesis testing procedure? |
1) State the hypothesis: both the null and the RQ 2) State the probability level (alpha): The probability that if an effect is found, this is due to sampling error 3) Consult a stats table to find a p-value to find out if it is less that the probability level 4) Conduct a statistical analysis 5) Accept or reject the null hypothesis |
|
One-Group t-test |
Is the sample similar or different from a defined known population? There are usually 2 alternate hypotheses so the probability level is set under a 2-tailed test. |
|
Two-independent groups t-test |
Are 2 samples different from each other? |
|
Two-dependent groups t-test |
Are 2 measures of the same sample differnt from each other? |
|
ANOVA |
Analysis of variance. Are more that 2 samples different from each other. Analyzes whether or not the means of several groups are equal and therefore generalizes the t-test to more than 2 groups |
|
One-way ANOVA |
With 2 groups this is the same as a 2-independent group t-test. |
|
Repeated Measures ANOVA |
Analysis when each participant is measured on 2 or more occasions. If there are only 2 measures this is the same as a 2-dependent groups t-test. |