2019: Things to think about in your CAFS Assessment
Before you begin your research, unpack the definitions and assumptions behind your question:
- What are 'electronic devices'? What is "screen time"?
- What counts as "responsible use", "excessive use" and "addiction"?
- Who created these definitions? What are the assumptions behind these definitions?
- What makes night time use a problem? Is it always a problem?
Work out the kinds of questions you need to answer in your secondary research:
- How much sleep do teens need?
- Does sleep need to be taken in a single 8-10 hour block? (research biphasic sleep patterns)
- What kinds of teens choose to use their devices late at night?
- What do teens do late at night on their devices?
- Are there any usage patterns (eg: gender, age, international students, self-esteem, SES, etc who use their devices in similar ways)?
- Are there specific triggers for late night use?
- Are there times when the pros of late night device use outweigh the cons?
When you've completed your secondary research, ask yourself:
- Has your research shown correlations (see the table below for which words to look for) ?
- Are there questions you couldn't answer in your secondary research that you still need to answer?
- Are there findings that you'd like to test in your primary research?
- Are there specific biases that have become apparent in the way the topic has been researched? How could you overcome them in yours?
- What kinds of measurement systems were used in the research (eg: the Problematic Mobile Phone Use Questionaire, the Pittsburgh sleep Quality Index)? Are they effective measurement systems?
- Will you have access to these measurement systems? If not, what effective measurement systems will you use?
When you've completed your primary research, evaluate what you've discovered about device use:
- Does your primary research match what you found out in your secondary research?
- Or does it challenge these findings?
- If it challenges these findings, what might have caused the differences (choice of subjects, definitions, methodology etc)?
- What things still need to be researched (things not covered in your primary or secondary research)?
- Are there things you could have done better in your research? How could it have been improved?
Correlation or Cause?
The language of an article may be persuasive, but if you look carefully at the conclusion, you can usually discover whether the article is describing two things which occur in the same research subjects (eg teenaged girls and high anxiety levels) or one thing which actually causes the next thing (eg low self esteem and increased anxiety).
This is important:
Research shows that many teenaged girls have high anxiety levels, but being a teenaged girl does not cause the anxiety - if it did all teenaged girls would suffer from high anxiety levels. The relationship between these terms is a correlation, not a cause. In a correlation, two things may be linked or they may have nothing at all to do with each other. They might even both be caused by a third thing that no one has tested yet.
In contrast, most people who have low self esteem have increased anxiety levels so we can assume that the low self esteem causes the anxiety. This is a cause and effect relationship.
Many media articles report cause and effect relationships when the research only shows a correlation. This is great for advertising or creating outrage because cause/effect relationships are very persuasive, especially if graphs are included. However, critical readers (which is what we hope you are becoming) will look at the language and go back to check the validity of the methodology used in the research - don't believe any article which doesn't at leave give details of the research and who conducted it so you can check for yourself. You can read more about this at Why correlation does not imply causation.
Because you are also clever researchers, we also hope you will use the correct words in your own research reports.
This is important:
Research shows that many teenaged girls have high anxiety levels, but being a teenaged girl does not cause the anxiety - if it did all teenaged girls would suffer from high anxiety levels. The relationship between these terms is a correlation, not a cause. In a correlation, two things may be linked or they may have nothing at all to do with each other. They might even both be caused by a third thing that no one has tested yet.
In contrast, most people who have low self esteem have increased anxiety levels so we can assume that the low self esteem causes the anxiety. This is a cause and effect relationship.
Many media articles report cause and effect relationships when the research only shows a correlation. This is great for advertising or creating outrage because cause/effect relationships are very persuasive, especially if graphs are included. However, critical readers (which is what we hope you are becoming) will look at the language and go back to check the validity of the methodology used in the research - don't believe any article which doesn't at leave give details of the research and who conducted it so you can check for yourself. You can read more about this at Why correlation does not imply causation.
Because you are also clever researchers, we also hope you will use the correct words in your own research reports.
Words that show correlation:
|
Words that show cause
|
Words that express inference (the researcher thinks it's a cause but isn't certain)
|