Analyze Data

The results are in! We had a total number of 890 kids participate in the Pack a Lunch with Punch activity. Is our hypothesis correct? Let’s take a brief look at the data. Keep in mind, this is just a summary. In science research, data can tell you a lot, and different people sometimes notice different important parts of the results.

• The results are in! We had a total number of 890 kids participate in the Pack a Lunch with Punch activity. View the results.

  Pizza is big with kids, especially younger kids and boys. 
  When looking at all kids attending the demonstration, pizza was the most favorite food, with boys (more than 50%) in both age groups making this choice.  Pizza was favored most by kids 9 years and younger. 
   
• Older kids like the wrap more than younger kids, and girls like it more too. 
  The wrap was more popular with kids 10 years and older than kids 9 years and younger.  More girls than boys chose the wrap in both age groups.  For kids 10 years and older, 73% of those choosing the wrap were girls.
   
• Boys’ preference differed from girls.
  Boys in both age groups preferred hot dogs more than girls did.
   
• Salad received a mixed response. 
  Boys 9 years and younger preferred salad about the same as boys 10 years and older.  Girls 10 years and older, however, preferred salad 2.5 times more than younger girls and more than boys in their age group.
   
• Younger kids prefer chicken pieces more than older kids. 
  Chicken pieces were favored 2 times more for kids under 10 years, with little difference between boys and girls. Though there is a small difference between boys and girls in preference for chicken pieces, it is probably too small to be important. Scientists call this “insignificant.”
   
• Older kids prefer fish and vegetables more than younger kids. 
  Kids 10 years and older preferred fish and vegetables 2 times more than younger kids.  While there was little difference between boys and girls 9 years and younger, 47% more girls 10 years and older preferred the dish over boys in the same age group.

Interpreting Data

While these results are very simple, and we can summarize them in just a handful of numbers, they actually help us address several questions.  Have you thought about what factors could influence these choices? 

• Do the choices of boys and girls differ?
• Do the choices differ when we compare younger to older kids?
• Is there a difference between younger versus older when looking at  responses from boys and from girls?

The last question refers to interactions between factors. These interactions can be critically important to recognize for many studies.

In our data, we see a difference in the percentage of children choosing simple versus complex food items when comparing younger to older kids; but upon further investigation, by looking to see if there is an interaction, we might discover that this apparent difference is partly or totally due to a change in the percentage of kids that are girls between the younger and older age groups.  In our case, for kids 9 years and younger, 42% are girls but, for kids 10 years and older, 55% are girls.  In this observational study, we cannot control this factor; and therefore we must be very careful in concluding that a difference in food choices between younger and older kids.  Because of interactions, this could really be an effect of age and gender, or it might be due to some other factor.

Interfering Factors

In a research study, like our demonstration study, it is important to try to understand what all of the possible interfering (also called confounding) factors might be.  If possible we need to try to take these into account in analyzing the data.  Here are some possible interfering factors:

• Day of the week
• Religious preference
• Diet preference (vegetarian or non-vegetarian, low carbohydrate, high protein, food sensitivities, etc.)

Even the presence of parents during the food choice can possibly influence the kid’s decision about their favorite food.  Understanding interfering factors is very important if we want to apply our results to a larger group or population.  For example, if we want to say that all kids age 9 and younger will like chicken pieces, we have to remember that there are interfering factors that may make our results only applicable to a smaller group of kids and not all kids who would eat lunch at our Clinical Center hospital.

Study Limitations

It’s important to know that even the most carefully designed studies have limitations, and we have to consider these in interpreting results.  Researchers must recognize and report these limits.  For example, what if, in our demonstration study, we had only asked girls about their food preferences?  In this case we would need to do two things.  First, we would have to make it very clear that only girls participated in our study.  Second, in reporting our results and what we think they imply, we would have to be careful not to expect the boys’ choices to be the same as the girls’ choices.  As our demonstration data shows, it might not make sense to use information about girls’ preferences to make a lunch menu specifically for boys.

A study or experiment usually helps to answer a particular question…it can also lead to ideas for further questions or studies.

Back to the main Pack a Lunch with Punch page

NOTE: PDF documents require the free Adobe Reader.