PSY123 Research Methods and Statistics Lecture 6 PDF

PSY123: Research Methods and Statistics Lecture 6 Important Announcements  Test 1 results available on iKamva on Monday, 29 August @ 8h30.  Each student may view his/her own scores on iKamva via the ‘Test and Quizzes” section and also via the ‘G...

PSY123: Research Methods and Statistics Lecture 6 Important Announcements  Test 1 results available on iKamva on Monday, 29 August @ 8h30.  Each student may view his/her own scores on iKamva via the ‘Test and Quizzes” section and also via the ‘Gradebook’.  End of 3rd term: 30 August, campus vacation period  Term 4 starts: Mon, 9 September (start of Lecture 7, Prof Serena Isaacs)  Recap: Last week we covered Step 3 Data Collection Levels of measurement Methods for gathering data What is “raw data”? This is the data which comes out of the data collection Levels of measurement of data 1. Nominal Describe categories. Also known as the e.g. the number of people who “naming” scale. helped/ not. a person's gender, ethnicity, hair color etc. are considered to be data for a nominal scale. 2. Ordinal Measurements that can be ranked in a She came in first, he came second, specific order, or put in position, but etc. Low, middle or high, usually used intervals are unknown/ have no meaning. in a Likert Scale response 3. Interval Measurements based on scales. Distance e.g. temperature between the scales are EQUAL and has The difference between 20 and 21 meaning. degrees is identical to the difference between 225 and 226 degrees 4. Ratio Is very meaningful, for it has an absolute This week we cover Lecture 6 Step 4 Data Analysis: making meaning of collected data with a focus on Qualitative Data Analysis (QDA) Analysis of quantitative data In quantitative research statistical analyses, involving analysis of numbers or figures, produce two broad types of data: descriptive and inferential statistics. Descriptive statistics refers to data which is used to describe or summarise the sample; e.g. number of individuals, the number of males and females or the age range of those in the sample. This information can be expressed in tables and graphs etc. Inferential statistics refers to data which can be generalised from the sample to the population. This data can be used to test hypotheses and therefore draw conclusions on the sample which can be applied to the population under study. Analysis of qualitative data 1. Qualitative data refers to non-numerical data. 2. Qualitative data often takes the form of interview transcripts. 3. Qualitative data also takes the form of documents. (EXAMPLE: government reports, books, newspaper articles), open-ended survey responses and the interpretation of images and videos. Analysis of qualitative data 1. The analysis involves sifting through transcripts in a systematic way so that conclusions may be reached. 2. There are several methods for analysis qualitative data which all revolve around the analysis of meaning. 3. Qualitative research is significantly different from quantitative research and therefore has its own verification processes.  What is qualitative data analysis? (QDA)  ANSWER: The range of processes where we move from: 1. the qualitative data that have been collected, 2. into some form of explanation, 3. understanding, or 4. interpretation of: the (a) people and (b) situations we are investigating. Use of Qualitative Research: a) understanding underlying reasons; opinions and motivations. b) provides insight into a problem. When analysing qualitative data: the researcher needs to look at (a) depth; (b) detail and (c) complexity CONTEXT is CRUCIAL: i.o.w. what connection to the social world does the participant have? EXAMPLE: Giving a student an “award” or “comment” [researcher needs to look at what was the student’s experience like at university?] was the experience negative/positive?  Transcription of Data Interviews and focus groups: 1) Are always audio-taped or audio-recorded. 2) Researchers must then transcribe these interviews into a written format (this is a very time consuming process, but necessary). Audio-tapes: 1) should be described word for word into an electronic document. REMEMBER: 1) You as the researcher should always insert comments, symbols that denote hesitation, pauses, laughs or sighs. 2) Transcribe entire interviews rather than what you want to select. 3) As the researcher transcribes the interviews, he or she has to make notes about possible interpretations.  Qualitative Data Analysis  The six most popular qualitative data analysis methods are: 1. Qualitative content analysis [read additional notes on iKamva] 2. Narrative analysis 3. Discourse analysis 4. Thematic analysis 5. Grounded theory (GT) [read additional notes on iKamva] 6. Interpretive phenomenological analysis (IPA) [read additional notes on iKamva)  2. Narrative analysis Narrative analysis is a technique that approaches the transcript as if it were a story following a particular sequence. 1. The word narrative is always interchangeable used with the word “story”. 2. A narrative always responds to the question: ‘And then what happened?’ 3. Narrative analysis is useful for a study that changes over time. Used to explain how participants in a study make sense and meaning of themselves and their actions. Example: A narrative of learners’ lives, establishes the importance of stories, provides an illustrative example of the analysis of an adult learner's story.  3. Discourse Analysis 1. Discourse simply emphasises the written or spoken language. 2. Discourse includes the relationships we have with other people, as well as with power. 3. Discourse is a process of reasoning about things in social realities. i.o.w. Reasoning about where you fit in in your own society. The researcher typically tries to draw attention to dominant meanings of particular phenomenon, as well as how such meaning may be ambivalent or contradictory 4. Discourse analysis entails analysing language in a political or social-cultural context. I. o. w. analysing language such as a conversation, a speech, a book or any text in the society we live in. 5. Discourse analysis wants to investigate the functions of language (i.o.w. what is language used for), how meaning is constructed in different contexts, which include social, cultural, political, Read additional and historical background of the discourse. notes on “Gender as Social 6. Discourse analysis is socially constructed which makes it unique. Construction”.  4. Thematic analysis (TA) Thematic analysis is the most commonly used method of qualitative data analysis (QDA). TA looks at patterns of meaning in the transcripts from interviews and focus group discussions 1) Audio-taped transcriptions are first broken down into units of meaning. 2) The researcher then uses a technique to place such units into categories. 3) Most common themes are systematically identified. 4) Discussions of large information sections are broken down into similarities – in other words, they are divided into themes. 5) From the themes: insight into particular issues being studied can be drawn. In other words, we make meaning of the content of these themes. Thematic analysis: Example 1  Thematic analysis Example 2 TRANSCRIPTS FROM THE INTERVIEWS POTENTIAL THEMES OR CODES  Step 4: Analysis of Qualitative data [QDA] In summary let us now look at the 3 types of analysis below: a. Transcription of data ONE of the most commonly used qualitative approaches (1) Narrative analysis Analysing how stories are told (2) Discourse analysis Analysing conversations and interactions (3) Thematic analysis Identifying themes and patterns (4) Grounded Theory Starting from scratch with a specific question to build a theory in response to that same question. (5) Interpretive Phenomenological This is about understanding people’s unique Approach (IPA) experiences of a phenomenon. b. Verification of qualitative data  What is data verification of qualitative research? 1. Verification is the process of checking and confirming. I.o.w the strategies are used to ensure reliability, validity and therefore the rigour of the study during the research process. 2. Several debates about the rigor of qualitative research around questions, viz: How can we trust the authenticity of qualitative research? How can we be sure that such research is reliable and valid? 3. There are many different perspectives on how to make sure that qualitative data is trustworthy and rigorous. 4. Correspondence checks is frequently used: this involves the use of: (a) colleagues and (b) other researchers to analyse the data independently. 5. Analysis is then compared with that of the primary researcher’s to check for correspondence. 6. Also: take the analysed data back to the participants to find out what they think of the analysis.  NOTE WELL: Always remember the goals, theory and method of study when interpreting the data 7. Final analysis: the researcher should provide enough information to allow others to assess: (a) the merits and (b) trust-worthiness of the work.  More types of verification of qualitative data How to verify qualitative data and promote rigour: 1) Triangulation: check out if stories are consistent- ask questions from two/ more different persons/sources. 2) Reflective Journal: A dairy kept by the researcher to provide personal thoughts and insights on what happened during the study, also noting personal biases,etc… 3) Data Saturation: This is a tool frequently used for ensuring that adequate and quality data are collected to support the study. 4) Member checking/ Respondent validation: Take data analyses back to the members of the group and confirm the data. 5) Self-disclosure (Reflexivity): Researcher should not be too critical. i.o.w. better to use more than one investigator. 6) Meticulous record-keeping: Demonstrating a clear decision trail and ensure interpretations of data are consistent and transparent.  HOMEWORK EXERCISE 1. Identify the six most popular qualitative data analysis methods. 2. What influences which method of data analysis a researcher chooses? 3. Explain the concept of social constructionism. 4. Why is gender a social construct? 5. What is triangulation? 6. Why is data verification important in qualitative research? 7. Identify and explain the two types of biases in qualitative research.  Homework exercise slide 2 8. Reflect on the following questions: 8.1 A research student wants to understand the personal experiences of young adults who were raised by a parent with a mental health problem. The researcher conducts 20 in-depth, semi-structured interviews. Which method of analysis would be most suited to this study? 8.2 TikTok is an emerging social media platform and a group of researchers (Li et al., 2021) want to investigate the video format, type, and content of official COVID-19 TikTokvideos. The researchers found that almost every video included a hashtag, and a higher number of hashtags was related to more likes. Video types included acting, animated infographic, documentary, news, oral speech, pictorial slideshow, and TikTokdance. What type of analysis has the researcher undertaken? 8.3 In a recent analysis of Addiction Science in South Africa, you come across two different articles: one written by a Health Professional working in the field of prevention and the other by a policy-maker. The Health Professional’s article emphasizes the science of prevention and primary health care and highlights the importance of substance use as a chronic disease, while the policy-maker’s article focuses on the economic impacts and necessary policy actions for addicts. What kind of analysis of these articles would you engage in?  REMINDERS 1. Special Test 1 dates: 9- 10 September (see iKamva announcement]  Please read for lecture 7  Lecture 7 starts from Monday 11 September 2023.  Please refer to the course outline for the specific page numbers