how to do a comparative synchronic analysis of correspondence analysis？

[florycst123]

Version number of KH Coder

khcoder3- mac version semi-manual installer 302-macdl

Your Qeuestion

Your Operating System

e.g. macOS sonoma 14.5, i3

Your Qeuestion
Hello, I currently have 3 correspondence analysis data for high-frequency words over x years. The problem is that the exported data from each correspondence analysis has different dimensions. For example, for the word "happy", the CSV table shows the total frequency in 5 years, along with total size and dimensions x1, x2, x3. If I want to perform a comparative synchronic analysis to quantify the similarities and differences among the three correspondence analyses in each year, how should I proceed?
another question：For a co-occurrence analysis of a given word, can KH Coder visualize the results over a specific time period, showing how the co-occurrences change over time?
Screenshots
If applicable, add screenshots to help explain your question.

What language of text are you trying to analyze?
ex. spanish etc.

[ko-ichi-h]

What happened to #1248 and #1277?

It's rude to abandon your threads and ask new questions.

[florycst123]

sorry I have just ended the previous two questions.

[ko-ichi-h]

Hello, I currently have 3 correspondence analysis data for high-frequency words over x years. The problem is that the exported data from each correspondence analysis has different dimensions. For example, for the word "happy", the CSV table shows the total frequency in 5 years, along with total size and dimensions x1, x2, x3. If I want to perform a comparative synchronic analysis to quantify the similarities and differences among the three correspondence analyses in each year, how should I proceed?

I don't understand what you want to do. Why are there three correspondence analysis results? Why not combine them into one analysis from the start?

another question：For a co-occurrence analysis of a given word, can KH Coder visualize the results over a specific time period, showing how the co-occurrences change over time?

In the co-occurrence network option screen, check "Explore changes in co-occurrence patterns" and select a numerical variable like "year".

[florycst123]

Since these three correspondence results are from three different corpora, I now want to compare the performance of the three results for each year to observe the differences among them in the same year

[ko-ichi-h]

I still don't understand.

What differences do you want to observe?

[ko-ichi-h]

What if we combined the three corpora into one, created a variable like "2020-Corpus1", and performed correspondence analysis using that variable?

This means creating one variable that contains both the period and the corpus name.

[florycst123]

Thank you very much for your suggestion. After trying it, I found that combining both the years and multiple corpora in one resulted in an excessive number of tokens in the visualization, making interpretation quite challenging. I may opt for manual coding as a more feasible approach.

[ko-ichi-h]

Next time, if you could attach the actual analysis results and write down what you are dissatisfied with, we may be able to give you more appropriate advice.

[florycst123]

I attempted your suggestion to use the co-occurrence analysis under the code section in KH Coder. I checked "Explore changes in co-occurrence patterns" and selected a numerical variable, such as "year." This interface allowed me to choose codes representing three different time periods. However, If I want to study a specific word, for example, person, and observe its co-occurrence words while analyzing the differences in these co-occurrences across three time periods in a single chart, how should I proceed?

[ko-ichi-h]

This interface allowed me to choose codes representing three different time periods.

Sorry, I can't understand this sentence.

Can you attach an example chart? And can you write down what you are dissatisfied with?

However, If I want to study a specific word, for example, person, and observe its co-occurrence words while analyzing the differences in these co-occurrences across three time periods in a single chart, how should I proceed?

Single chart... Hmmm, is single table ok?

[florycst123]

I apologize for not expressing myself clearly. I want to study the top 20 co-occurring words of person in different periods to examine the common and unique co-occurring words associated with person across these periods. My variables are Period 1, Period 2, and Period 3. While I can obtain the co-occurring words for each period (variable), how can I ensure that these co-occurring words are all related to the word person? can I get a visualization of this？

[ko-ichi-h]

1. Go to [Tools] [Words] [Word Association]
2. Enter "<>PERIOD-->period 1" person , hit [Run] button, copy results to Excel
3. Enter "<>PERIOD-->period 2" person , hit [Run] button, copy results to Excel
4. Enter "<>PERIOD-->period 3" person , hit [Run] button, copy results to Excel
5. Create a table like this: https://speakerdeck.com/koichih/kh-coder-tutorial?slide=22 (Create a table like the one on the right side of this page. Ignore the instructions on the left side.)

You may list up top 20-30 words for each period, 10 is too short maybe.

[ko-ichi-h]

You need to enter like:

"<>PERIOD-->period 1" person

You need the double quotes.