Which statement correctly contrasts reliability and validity in qualitative coding?

Which statement correctly contrasts reliability and validity in qualitative coding?

• A. Reliability is about consistency of coding across raters; validity is about whether the codes capture the intended themes.
• B. Validity is about consistency; reliability is about accuracy.
• C. Reliability and validity are the same in qualitative coding.
• D. Validity concerns researcher bias only.

Answer: (A)

In qualitative coding, you’re evaluating two different aspects of how codes reflect the data. Reliability is about consistency across coders or across coding passes—if multiple researchers code the same interview in the same way, reliability is high. Validity is about whether the codes actually capture the themes and meanings the data are intended to convey, meaning the coding scheme aligns with participants’ expressed ideas. So the statement that reliability concerns consistency across raters and validity concerns whether the codes capture the intended themes is the best description. For example, two researchers applying the same codes to a transcript shows reliability, while showing that those codes accurately reflect what participants are communicating demonstrates validity. The other ideas mix up these notions: reliability isn’t about accuracy or truth in the sense of correctness of content, and validity isn’t just about researcher bias. They’re distinct aspects—consistency versus fit to the data’s themes.

In qualitative coding, you’re evaluating two different aspects of how codes reflect the data. Reliability is about consistency across coders or across coding passes—if multiple researchers code the same interview in the same way, reliability is high. Validity is about whether the codes actually capture the themes and meanings the data are intended to convey, meaning the coding scheme aligns with participants’ expressed ideas.

So the statement that reliability concerns consistency across raters and validity concerns whether the codes capture the intended themes is the best description. For example, two researchers applying the same codes to a transcript shows reliability, while showing that those codes accurately reflect what participants are communicating demonstrates validity.

The other ideas mix up these notions: reliability isn’t about accuracy or truth in the sense of correctness of content, and validity isn’t just about researcher bias. They’re distinct aspects—consistency versus fit to the data’s themes.