MAIS 640

For this stage of the course, I have started working more directly with grounded theory data collection and analysis. My current research question is: How do people respond to expressions of loneliness and emotional distress in public social media discussions?

I chose this topic because I am interested in how people communicate care, support, discomfort, advice, or avoidance when someone expresses emotional pain online. I am not trying to diagnose anyone or decide whether a response is good or bad. Instead, I am trying to understand what people are doing with their words in the conversation.

​

This stage has helped me see that grounded theory is not only about collecting data and coding it. It is also about slowing down, staying close to the data, comparing examples, and being aware of my own assumptions. Charmaz explains that grounded theory is both systematic and flexible, and I am beginning to understand what that means in practice (Charmaz, 2014). There is structure, but there is also uncertainty. I have to follow the data rather than forcing it to fit what I already think (something I still struggle with).

​

Choosing and Narrowing My Data Source

For my data collection, I decided to use public Instagram posts and comment threads. I chose Instagram because it is a social media space where people often share emotional experiences, reflections, and personal struggles. I also chose Instagram because I am more familiar with it than TikTok, and it seemed easier to search for what I was looking for than on Facebook. If I used Facebook, I was not sure it would be as easy to find relevant public posts or stories connected to my topic. Instagram hashtags gave me a more practical way to find public posts about loneliness and emotions.

​

To find posts, I am using a few public Instagram hashtags related to loneliness, emotional distress, and emotional well-being. The ones I have been using so far are

​

• #loneliness,

• #mentalhealth

• #emotionalwellbeing

​​

I chose these particular hashtags because they help me find public posts that connect to my research question. I am keeping the search narrow so the data stays manageable. Instead of looking at every possible hashtag or topic, I am focusing on a few that are closely related to people who feel lonely or experience emotional distress. At the same time, I am not fixed on these hashtags. I am open to changing or adjusting them depending on what I find in the data. I don’t mind adjusting, which is probably something I wouldn’t have said 2 weeks ago!

​

My professor’s feedback on ePortfolio #1 also reminded me to consider theoretical sampling as I move through the data. I understand this to mean that I should not just collect data once and stop. I need to pay attention to what the first data shows, what questions remain unanswered, and whether I need to look at different hashtags, posts, or comment threads to better understand the patterns emerging.

​

I decided not to use information from people I know. This felt important for ethical reasons. Even when social media posts are public, they still involve real people sharing real emotions and experiences. This is, so far, the only clear limit I set. I may find other limits as I continue to collect data, but so far, that is the only ‘hard stop’ for me.

​

​Collecting the Data

My data collection has focused on public Instagram posts that express loneliness, sadness, emotional distress, or a sense of ‘not being okay’. I then looked at the public comments on those posts.

​

I am paying attention to the words people use, the tone of the response, and what the response seems to do in the conversation. For example, some comments validate the person’s pain. Some offer advice. Some encourage the person to seek help. Some share a similar experience. Some try to make the person feel better by saying that things will improve. Others seem to move away from the discomfort of the original post.

​

At this stage, I am not trying to collect a huge amount of data. I am trying to collect enough to begin seeing patterns and to practice initial coding, memo writing, and constant comparison. This feels appropriate because the goal is not to make a final claim about all social media responses. The goal is to learn how grounded theory analysis works by staying close to the data and building ideas from there.

​

Initial Coding

Initial coding has been one of the most useful parts of this stage, but it has also been challenging. I am trying to code line by line and use active words that describe what people are doing in their responses. Charmaz encourages researchers to use active coding because it helps keep the analysis focused on process and action (Charmaz, 2014).

​

Some of my early codes include:

• validating pain

• offering advice

• minimizing distress

• sharing similar experiences

• encouraging help-seeking (counselling)

• offering reassurance

• normalizing

​​

When I first started coding, I noticed that I wanted to label many comments as simply supportive. However, as I looked more closely, I started to see that support can work in different ways. For example, one person wrote, “I hear you, and that sounds really hard.” Another person wrote, “You just need to stay positive.” Both comments may be intended to help, but they do different things in the conversation.

​

This is where initial coding has helped me slow down. Instead of asking, “Is this supportive?” I am asking, “What is this response doing?” Is it validating the person? Is it redirecting the person? Is it minimizing the distress? This feels like an important change in my thinking.

​

Constant Comparison

Constant comparison has become an important tool in Stage II. As I compare comments, I am starting to see differences between responses that seemed similar at first.

​

For example, advice and validation can both seem supportive, but they do different things. Validation stays focused on the person’s feelings. Advice moves the conversation toward what the person should do next. Advice can be helpful, but it can also distract from what the person is feeling.

​

I am also comparing comments that try to reassure the person. Some reassurance feels helpful, but some can make the person’s pain seem smaller. For example, saying “You are not alone” may help someone feel understood. But saying “everyone feels this way” can make their feelings seem less serious. Small wording/language changes can make a big difference.

​

This is where constant comparison helps me. It helps me ask simple questions about each response. Is the person validating the feeling? Are they changing the topic? Are they making the pain seem smaller? Are they trying to fix the problem?

​

I am not trying to decide if the response is good or bad. I am trying to understand what the response is doing in the conversation.

I also noticed that I sometimes compare the data to my opinions. When I read certain responses, I think, “That is probably what I would say.” For example, I am drawn to responses that try to reduce awkwardness or make the person feel better quickly. That does not mean I should code the data based on my own reaction, but it does help me notice how my own habits may shape what I see. I know my reaction reflects neoliberal culture, and I do not want to force this idea onto the data, but noticing this reaction is a good step in the right direction.

​

Early Patterns I Am Noticing

One early pattern is that many people respond to emotional distress by offering comfort quickly. This can include reassurance, positivity, advice, or encouragement. These responses may be kind, but they may also move the conversation away from the pain that was expressed.

​

Another pattern is that some comments validate the person’s feelings. These responses show that the person’s emotions are taken seriously. People make comments like, “I hear you,” “That sounds hard,” or “You are not alone.” These comments let the person share their feelings/create space without trying to fix them right away.

​

I am also noticing responses that seem to minimize emotions. These comments could potentially be trying to help, but they can make the person’s pain seem less serious. For example, I have come across comments like “Everyone feels this way” and “Things will get better.” These comments may be meant kindly, but they can also make it harder for the person to share how they really feel. In ‘real life’, these kinds of comments often make me uncomfortable because they can feel dismissive. Noticing my own reaction is important because it reminds me to stay close to the data and not assume the intention.

​

Another pattern is that some responses move quickly into advice. These comments tell the person what to do, like “stay positive.” At first, I might have seen these comments as supportive. But through constant comparison, I am starting to see that advice is different from validation. Advice can be helpful, but it can also move attention away from the person’s feelings and toward fixing the problem.

I am also noticing that some people respond by sharing their own similar experiences. These comments can create a connection, but they can also shift the conversation's focus. Sometimes sharing a similar experience seems to say, “I understand.” Other times, it may shift attention away from the original post and toward the commenter’s own post.

​

A possible emerging idea is that many responses are not only about supporting the person in distress. They may also be about managing emotional discomfort in the conversation. When someone expresses loneliness or emotional pain publicly, it can make others uncomfortable. Some responses seem to stay with that discomfort, while others try to move past it quickly.

​

Memo Writing

As I’ve mentioned previously, memo writing felt/feels uncomfortable because it seems less structured than the kind of writing I usually do. I like knowing where something is going. With grounded theory, I am learning that memo writing does not need to be perfect. It is a place to think on paper.

​

This is probably not exactly how memo writing is supposed to look, but I bought a notebook, and that is where I have been writing my thoughts as I work through the data. There is not as much writing in it as I thought there would be, and I think part of that is because I am comparing it to the dramatic journals people write in shows. It is not full of polished entries. It is more like quick thoughts, small observations, questions, and reminders about what I am noticing in the data. Sometimes I am not sure what to write because I felt like I was already thinking it, so I wondered why I had to write it down. Then I remembered Charmaz’s point that memo writing is part of the analysis, not just a record of what I already know. So, I have been trying to force myself to write something, even when it feels awkward.

​

Reflexivity

Reflexivity is important in this project because I am not separate from the topic. My background in social services means that I often think about loneliness, emotional distress, wellness, and access to support. This experience may help me notice important details in the data. It may also shape how I interpret what I see.

​

For example, because of my work, I may be more likely to notice whether a response validates someone’s pain or directs them toward support. I may also be sensitive to comments that seem to minimize distress. This can be useful, but it can also create bias if I assume too quickly what a comment means. I need to keep asking myself questions such as: What am I noticing? What am I assuming? Am I staying close to the words in the data? Am I judging the response? These exact questions are on the 1st page of my notebook.

​

​

​

​

​

​

​

​

​

​

​

​​This has also made me think about my own responses to distress. Some comments stand out to me because they sound like something I might say. I may be drawn to responses that take the awkwardness out of the moment. It's an interesting reflection because it shows how my own experiences and habits are part of the research process. Reflexivity helps me see this instead of pretending it is not there.

​

Challenges So Far

One challenge is avoiding assumptions concerning intention. I cannot know what the commenter meant unless they say it directly. Someone may believe they are being encouraging, even if the response sounds dismissive. This means I need to code what the comment does, not what I think the person intended.  This is more difficult than it sounds.

​

Another challenge is deciding how specific my codes should be. Some comments could fit more than one code. For example, a comment could validate the person and also offer advice. I am learning that this is not necessarily a problem. It may actually be part of the analysis. A response can do more than one thing at the same time.

​

Another challenge is moving ahead too quickly. I am anxious to get to the analysis stage because that is my personality. I like organizing ideas and seeing where they go. However, this stage has reminded me that the slower work matters. Data collection, initial coding, constant comparison, and memo writing are helping me build a stronger foundation before moving into an  analysis.

I am also finding it difficult to work with emotional data. Even though the data is public, it still represents real people. I want to be respectful and careful. I do not want to overstate what I can know from a comment thread, nor turn emotional expressions into simple categories without recognizing their complexity.

​

What I Am Learning About Grounded Theory

This stage has helped me understand grounded theory more practically. In Stage I, I understood the concepts of coding, constant comparison, and memo writing more theoretically. In Stage II, I am beginning to see how they work together.

​

I am also learning that grounded theory requires patience, which I don’t have a lot of! It is not about proving what I already think. It is about allowing patterns to emerge through close attention to the data. This can be uncomfortable because I do not yet know where the analysis will go. At the same time, that uncertainty is also part of the method.

​

Next Steps

My next step is to continue reviewing Instagram comment threads, use hashtags and continue coding. I want to keep comparing responses across posts to see whether the same patterns persist.​ I also want to continue writing memos, especially about the difference between validation, advice, reassurance, and minimization. These categories seem important, but I need to keep testing them against the data.

​

As I move toward the next stage, I want to begin thinking more carefully about which codes may become early categories. I am especially interested in the idea that public responses to distress may involve both care and discomfort. Some people stay with the person’s pain, while others try to move the conversation toward reassurance, advice, positivity, or action.

​

One last thing…

GT has ruined Instagram for me 😊! What used to be mindless scrolling has now turned into me silently coding strangers in the comment section. I will be on Instagram on my own time and suddenly catch myself thinking, “Is this person validating, minimizing, redirecting, or offering advice?” Apparently, even my relaxing is now data collection. As funny as that is, it has also helped me see how grounded theory is changing what I notice in everyday interactions, not just in the data I am ‘formally’ collecting.