Qualitative Data Analysis PDF

CW-HANDOUT1 RSRCH1-HANDOUT8 Topic: Qualitative Data Analysis - Archetypes I. Introduction After transcribing, cleaning, and coding the qualitative data, we can finally analyze the data. Remember that the data analysis method that you will use will depend on your research problems and goals. With...

CW-HANDOUT1 RSRCH1-HANDOUT8 Topic: Qualitative Data Analysis - Archetypes I. Introduction After transcribing, cleaning, and coding the qualitative data, we can finally analyze the data. Remember that the data analysis method that you will use will depend on your research problems and goals. With that in mind, we will be discussing four qualitative data analysis methods: Archetypal Analysis - usually used for research done for the purpose of narrating a story or an experience. Content Analysis - usually used when analyzing words and phrases, and interpreting their meanings. Thematic Analysis - usually used when the purpose of the research is the creation of themes and patterns that can be used to gain insights and concepts related to a phenomenon Semiotic Analysis - usually used to analyze symbols and their cultural context and meanings. The first kind of qualitative data analysis that we will discuss is Archetypal Analysis. II. Archetypes Archetypes are symbolic forms of stories, persons, places, or images that have been accumulated from human experience throughout history. This kind of analysis is almost always used in literary works (fictional works) in order to give them universal acceptance, as these archetypes tend to transcend culture and race. The following are the most common archetypes: Hero - a character who predominantly exhibits heroic qualities, such as goodness and bravery, in order to fight evil and restore peace to society. Mother - may be: Fairy godmother - guides, supports, and directs the hero Mother earth - contacts people to support the hero and give spiritual and emotional nourishment Stepmother - treats the hero harshly and unfairly Innocent - the character who is inexperienced and possesses many weaknesses, but is well-liked because of the trust he gives others. He usually cowers and find safety from others. However, he will overcome these weaknesses and become someone who is strong. Mentor - the character that protects the hero and gives wise advice and counsel. He also trains the hero in order to achieve his goals. Doppelganger - the character that is a duplicate or evil shadow of the hero, which shows the evil side of the hero’s personality Scapegoat - a character that is false blamed for all the misfortunes that occurs Villain - the character who opposes or hinders the hero from achieving his goals, and is usually the person to be defeated in order to defeat the evil and restore peace. III. Using Archetypal Analysis Archetypal Analysis is best used in qualitative research focused on the creation of stories, such as in Narrative Research. This is because archetypal analysis allows for the creation of stories the readers can relate to, since archetypes provide universal acceptance. In order to use archetypal analysis, the following steps must be done: In the coding preparation, certain codes must be identified as pertaining to specific archetypes. This will make the archetypal analysis easier. Take note that most probably, the group which is the focus of the study is any of the following: hero, the innocent, or the villain. Identify which archetypes are present in the narrative. When creating a narrative, these archetypes must be identified and their stories clearly narrated. For example, if you want to create a narrative about frontliners during the pandemic, you must identify the archetypes in their stories. The doctors who fight the covid pandemic may fall under the hero archetype. Thus, as a researcher, you must describe their heroic characteristics and how they are shown in the narrative. You must also find the evil that they need to vanquish. You must also find the other archetypes. For example. In this case, the poor Filipinos who contract the virus and are therefore blamed for the spread of the disease may be viewed as the scapegoat. You, as a researcher, must find their stories and explain how and why they became the scapegoats, and who pointed to the poor as such. Your end goal when doing archetypal analysis is to find common archetypal symbols in data that you can use to weave a narrative of what is going on in a phenomenon. CW-HANDOUT1 RSRCH1-HANDOUT9 Topic: Qualitative Data Analysis - Content Analysis I. Introduction The next qualitative data analysis that we will discuss is Content Analysis. Content analysis is the systematic process of categorizing textual data into clusters of categories. The following must be remembered when doing content analysis (though the same may be applied to other qualitative data analysis): The data is open to subjective interpretation (Remember: Qualitative Research assumes that there are multiple realities). The data may reflect multiple meanings. The data is context-dependent. II. Conducting Content Analysis The following techniques may be used when doing content analysis: Identification of Patterns - after the data is coded, you can identify patterns by doing the following: Take note that certain respondents share the same responses to questions. Looking at similarities among these respondents, you can find a pattern. Take note that certain groups of respondents also have different responses to questions, finding the underlying reason for this may yield a pattern. These patterns may be another variable (sex, gender, age, etc.) or factor (geographical location, being born before a specific notable event, etc.). Also remember that patterns must be investigated and evidence must be gathered in favor of it as the analysis progresses. Clustering - Qualitative Data that falls under the same code may be clustered, and it may be analyzed and explained as a cluster. Metaphors - Metaphor Creation is the method of categorizing data through their similarities while ignoring their differences. It is usually done through the use of metaphors (archetypes may be used here). This may be used in connection with Identification of Patterns, where a data set coming from the same metaphor may be analyzed in the same way as clustering. Counting - While numbers are usually ignored in qualitative research, the frequency of a word being mentioned by different respondents may mean something. Thus, creating a frequency table may be a basis of analysis. Making Contrasts and Comparisons - Just noting similarities and differences between the data may yield an analysis, especially if such comparisons are repeated. Explaining why such similarities and differences exist is also another basis of analysis. Partitioning - this is the reverse of clustering. Using partition, a dataset thought to be belonging to one code or cluster is partitioned or divided into new codes. III. Tips on Conducting Content Analysis The following are tips that must be remembered when conducting Content Analysis: Remember that your research goal, research question, research design, and qualitative data gathered will determine which Content Analysis technique/s you will use. You are not required to use all of the techniques. Use only what is necessary in order to answer the questions of your research. Let the data speak. As much as possible, all of your analyses must come from the data. Also, as much as possible, NEVER ASSUME. As much as possible, all of the transcripts related to the research must be analysed using the techniques mentioned above. RSRCH1-HANDOUT10 Topic: Qualitative Data Analysis - Thematic Analysis I. Introduction The third Qualitative Data Analysis is called Thematic Analysis. It is similar to content analysis, but instead of categories, qualitative data is grouped using themes. Themes are usually experiences, meanings, or perceptions of respondents as regards a phenomenon. The main difference between Content Analysis and Thematic Analysis is the role of the research problem. In Content Analysis, the research problem guides the process as well as the techniques to be used in analyzing the data. In Thematic Analysis, the data gathered from different respondents are grouped into: Recurring meanings, experiences, and perspectives relating to the same phenomenon Differing meanings, experiences, and perspectives relating to the same phenomenon Manifestations of these meanings, experiences, and perspectives These recurring and differing meanings, experiences, and perspectives relating to the same phenomenon can be grouped into themes. These themes are then divided into sub-themes in order to better understand the phenomenon. This better understanding can be accomplished by analyzing these sub-themes and how they differ with other subthemes. These sub-themes can be further divided into subgroups, if necessary. For example, four different respondents who are parents are asked as regards their perception on the use of AI for academic requirements and how it affected their child’s learning. Here are some of their responses: Nakakaawa yung anak ko kasi di makasabay sa gamit ng AI na yan. Siyempre di naman lahat laging may gadget o internet. Parang unfair sa amin, di namin kaya yan e. Yung mga kaklase mataas lang grades kasi naka-AI naman, siya di ganon kataas kasi sariling aral lang niya talaga. Nakaka-disappoint lang. Dahil diyan parang di na natututo anak ko, dependent na lang sa AI. Lahat pala pwede na sa internet. Baka maging robot na anak ko niyan, di na siya gagamit ng utak e. HAHAHAHA! Ang bobo ng naka-isip ng AI. Edi lalong tinamad mga bata. Mamaya mali pala tinuturo na niyan. Using thematic analysis, we can create the theme ACADEMIC INJUSTICE. The terms “‘parang unfair sa amin”, “nakaka-disappoint”, “di na siya gagamit ng utak”, and “ang bobo ng naka-isip ng AI” may reflect different experiences, but all of them fall under the same theme. All of the respondents find injustice with the use of AI for the academic requirements of their children. Furthermore, since they all fall under the same themes, they can be divided into sub-themes. A possible sub-theme is the manner by which they vent out their disappointments: The first one may be frustrated, the person is frustrated because they have no choice or chance to adapt to the modern system. It frustrates them that they are falling victim to an unjust educational system. The second one may fall under indifference. It appears that everything that is going on is an injustice since AI seems to hinder the child from learning naturally. The third one uses sarcasm, and hides his feeling of injustice behind jokes as AI seems not to bring justice to improve the capacity to learn of a child. The fourth one is angry, and uses derogatory words to express his disgust to the new AI system. Take note, however, that the creation of themes and sub-themes is not as easy as what the example made it appear to be. Each theme and sub-theme must be thoroughly checked in order to ascertain its correctness. The researcher must check the transcripts to determine inconsistencies and erroneous analysis. In the same way, other parts of the transcripts may enforce your themes, making your analysis stronger and more correct. II. Doing Thematic Analysis The following are steps that may help researchers in doing thematic analysis: Re-read the transcripts. This will help us understand the answers of the respondents in a more goal-oriented manner (because now you will be looking for meanings, experiences, and perspectives as regards a phenomenon). Review the CODES. Since thematic analysis is actually part of Selective Coding, reviewing the codes can help us identify themes. Usually recurring codes related to experiences, meanings, and perspectives are possible themes. List all possible themes. Then, label codes or transcript parts that are part of these themes. Once you identify themes, divide them into sub-themes using data provided in the transcripts. Make sure that these sub-themes are backed by data. This will help in the presentation part. Review the themes and sub-themes, and determine whether there are inconsistencies or errors pertaining to them. You can also check for overlaps. RSRCH-HANDOUT11 Topic: Qualitative Data Analysis - Semiotic Analysis I. Introduction The last qualitative research that we will be discussing is Semiotic Analysis. It is basically the analysis of data using semiotics. Semiotics is the study of signs and their meaning in society. A sign is something which can stand for something else – in other words, a sign is anything that can convey meaning. The following are examples of signs: 1. Words 2. Drawings 3. Photographs 4. Street signs 5. Modes of dress and style II. Branches of Semiotic Theory There are primarily two main schools of thought as regards semiotic theory. The first branch of semiotic theory provides that all elements of language are taken as parts of the larger system of language in use, and in fact all components of language are defined not in terms of some absolute standard, but by their relations to other components within the overall system. The key component of the sign relation was the dyadic relation of the signifier (or the sign proper) to the signified (or the concept of something which the sign triggered in some sign user). The second branch is based on the Aristotelian notion of potency and act. However, this characterization was expanded into a triadic model: potency, act, and relation. There are concepts related to the second branch of semiotics: Firstness - Those aspects of reality which deal with and characterize pure potency. Firstness deals with (among other things) issues of possibility. Secondness - Those aspects of reality which deal with and characterize pure action. Secondness deals with such things as brute force, pure reaction, and pure awareness that something is happening here and now, without knowing or understanding what it is that is happening. In other words, Secondness is the pure action-reaction relationship. Thirdness - Those aspects of reality which deal with and characterize relation and lawlike actions and situations. Thirdness deals with such issues as rules, laws, and habits. Anything symbolic, including language and sign systems in general, are real as Thirds. But any system of Thirds embeds and contains prior systems and components of Firsts and Seconds. We will be primarily using the first branch in doing semiotic analysis, but the second branch can help in identifying meanings in signs and symbols. III. Doing Basic Semiotic Analysis The following are steps that a researcher can do in order to do a Semiotic Analysis. Identify a specific lens (or perspective) that you want to use in order to view the objects. This can make the analysis more grounded. For example, a political advertisement usually targets a specific audience, and only that audience can truly understand the signs. This is grounded on the concepts of the Signified and both Secondness and Thirdness. Look at the objects and search for signs. Remember that anything that conveys meaning may be considered as a sign. List all possible signs in the object you are analyzing. This is based on the concept of the Signifier. Note that there are three possible forms of the Signifier: Icons - physical representations Symbol - non-physical representations Index - cross-referential representations For each sign, identify possible meanings. This is grounded on Firstness. Pure concepts do not exist by themselves. Thus, they are manifested in objects. There are three possible meanings: Denotation - the literal meaning Connotation - what the sign means from the point of view of the person looking at it Myth - symbolism based on socio-cultural contexts RSRCH1-HANDOUT12 Topic: Qualitative Data Interpretation I. Introduction Let us have a brief review of what we discussed so far. After gathering qualitative data, researchers must do the following processes: Transcribe the Data. Clean the Data. Code the Data using the following: Open Coding Axial Coding Do Selective Coding by choosing at least one of the following: Archetypal Analysis Content Analysis Thematic Analysis Semiotic Analysis Interpret the Data (possibly using the Research Framework). Present the Data. Now that we finish analyzing the data, the next process is interpreting and presenting it. II. Interpreting the Data Interpreting the data means that all the results of the qualitative data analysis process is summarized and synthesized. The main purposes of this process are the following: a.) Answering the questions enumerated in the Statement of the Problem is a very brief and concise manner. b.) Making the explanation of the aforesaid answers to the Statement of the Problem very understandable for the readers of the research. III. Steps in Data Interpretation The following are guides that must be following when interpreting qualitative data: Review the archetypes, analyses, themes, or sign representations you made from the qualitative data. Answer the questions in the Statement of the Problem based on the results of the data analysis. These answers must be straightforward. It must be brief and concise. AFter giving a straight answer to the question, enumerate points or important findings that serve as evidence or basis for your answer. Explain these points or important findings one by one. When explaining EACH POINT you can do the following tips: You can quote answers or examples given by your respondents. You can compare or contrast your points or findings with existing literature or research. You can analyze the qualitative data using the paradigm explained in your research framework. You can compare or contrast conflicting or opposing answers from your respondents. Then, explain how said differences are resolved. Furthermore, depending on the data analysis (or selective coding) you chose, you can do the following: If you choose Archetypal Analysis, you can add interesting stories where these characters demonstrate their archetypal characteristics. If you choose Content Analysis, you can emphasize the consistency of the answers and meaning attributed to a specific phenomenon. You also note any deviation to any established pattern. Also, you can give meanings to frequent responses. If you choose Thematic Analysis, you can identify common themes and connect this theme to the research question. If you choose Semiotic Analysis, explain the meaning of the signs by connecting it into a paradigm or by connecting the various meanings into a one coherent explanation. You can also create graphical representations in order for the readers of the research to easily grasp the result of your study. The aforementioned are the primary steps needed in doing data interpretation. Remember that these are merely guides and if you thought of something that might better present and interpret the results of your data analysis, then you are free to do so. RSRCH1-HANDOUT13 Topic: Quantitative Research Design (Descriptive Research) I. Introduction CONGRATULATIONS, researchers for God and County! You completed all lessons related to Qualitative Research. Now, we will be focusing on Quantitative Designs Methodologies. We will start by discussing Chapter 4: Research Methodology. Chapter 4: Research Methodology is the section of the Research Paper which: Explains the methods used Explains why they are used Identify and describes concepts used in the paper Explains how these concepts informed or influenced study Explain why the researcher adhered to these concepts Please take note that this part of the research paper can be applied to both Qualitative and Quantitative Research. II. Concepts related to the Quantitative Research Methodology The following are concepts related to the Chapter 4: Methodology: Method is the technique used to gather and generate data. Research Design is a plan which structures the study to ensure that the data collected and generated will contain the information needed to answer the initial inquiry as fully and clearly as possible. Population is the sum of all the units of analysis. Sampling is the process of choosing a representative part of the population under study. Statistical Treatment is the method in which statistical techniques are incorporated and applied in order to solve the research problem. These statistical techniques can be said to be the QUANTITATIVE COUNTERPARTS of qualitative data analysis techniques such as Content Analysis, Thematic Analysis, etc. Level of Measurement is the way chosen by the researcher/s in order to measure the variables. III. Three Types of Quantitative Research Designs There are the three primary types of Quantitative Research Designs. The first one Descriptive Research will be discussed in this lesson. The other two will be discussed next lesson. The three quantitative research designs primarily utilized in GFA SHS are: Descriptive Research Associational Causal-Comparative Research Associational Correlational Research IV. Descriptive Research Descriptive studies describe a given state of affairs as fully and carefully as possible. The most common descriptive methodology is the survey, as when researchers summarize the characteristics (abilities, preferences, behaviors, and so on) of individuals or groups. Example of Descriptive Research: An Exploratory Study on the Level of Happiness of Grade 11 Students in GFA SHS A Study on the Need to Belong of Grade 12 Honor Students in Taytay, Rizal The goal of these studies are just to describe the variables as it exists in the studies’ units of analysis. The first study describes how happy Grade 11 students in GFA SHS are, while the second study describes the need of Honor Students in Taytay to belong in a social group. The description of phenomena is the starting point for all research endeavors. As such descriptive studies usually start a chain of research regarding a phenomenon. This makes Descriptive research, in and of itself, not very satisfying. This is because the researchers will not really be able to understand the roots of the phenomenon, but just describe it. RESEARCH-HANDOUT14 Topic: Causal-Comparative and Correlational Research Designs I. Introduction Last lesson, we discussed Chapter 4: Research, as well as the different concepts related to it. Also, we discussed the first type of Quantitative Research Design: Descriptive Research. Now we will be discussing the two Associational-type research designs: Causal-Comparative and Correlational Research Designs. However, before we discuss these two research designs, we must understand the concepts of Causal Effect and Correlation. Causal Effect or Causation occurs when a variation in one phenomenon, an independent variable, leads to or results, on average, in variation in another phenomenon, the dependent variable, without any unintentional interference from a third variable. Correlation, on the other hand, is a type of non-experimental research in which the researcher measures two variables and assesses the statistical relationship (i.e., the correlation) between them with little or no effort to control extraneous variables. Remember that correlation does not mean causation. Just because two variables are correlated does not mean that one is the cause of the other. If two variables, however, have a causal relationship, then they are correlated. II. Associational Research Compared to Descriptive Research, the purpose of Associational Research is not just to describe a phenomenon but to determine possible relationships related to it. The purpose of determining said relationships are the following: Investigating such possible relationships will enable researchers to understand phenomena more completely. Identifying such relationships enables researchers to make predictions. III. Similarities and Differences between Causal-Comparative and Correlational Research Causal-comparative and Correlational research methodologies are the principal examples of associational research. Both types of research study whether two or more variables are related in a statistical context. Thus, if their values change so that as the value of one variable increases or decreases so does the value of the other variable (although it may be in the opposite direction), they are deemed to be related. Causal-comparative Research is a research design which aims to determine whether causation exists between two variables. This is accomplished through comparison of two groups of people: one possessing the quality of the variables being tested and another group which does not possess such quality. If there is a statistically significant difference between the groups, then it can be said that the difference is the cause of the other variable. Example: A Comparative Study on the Fear of Happiness between Honor and Non-honor Students in GFA SHS In this example, the goal of the study is to determine whether being an honor student is the cause of fear of happiness. The researcher will compare the fear of happiness (dependent variable) of both honor and non-honor students. Then they will determine two things: Whether the two groups are different when it comes to fear of happiness If they are different, whether this difference is statistically significant and enough to determine causation Correlational Research is a research design that aims to describe the size and direction of a relationship between two or more variables. This relationship may be: Positive - where an increase in one of the variables results to an increase in the other variable Negative - where an increase in one of the variables results to a decrease in the other variable Non-existent - where there is no relationship between the variables, or the relationship is deemed weak or insignificant by statistical treatment Example: A Study on the relationship of Happiness and Academic Performance among SHS Students in Taytay, Rizal In this study, the goal of the study is to determine whether there is a relationship between being happy and having high grades. The possible results are positive (as happiness increases, grades increases), negative (as happiness increases, grades decreases), or non-existent (there is no significant relationship between happiness and grades)