Candle question?

[deleted-80358]

I am doing a science project on which colored candle melts the fastest.
for my hypothesis I put "If the lighter color of the candle is related to the pigment in the candle, then the more evenly the pigment distribution is, the faster the candle will melt"
Does my hypothesis make sense? And does the pigment in the candle or the distribution of the pigment affect how fast the candle burns?

Please help me explain why or why not... if not, then I need a new hypothesis

[donnahardy2]

Hi,

For your science project, the hypothesis should be based on the science behind the experiment. In the case of candles, they burn because they are based on paraffin. Here’s a website that describes some of the complications of combustion chemistry:

http://www.newton.dep.anl.gov/askasci/c ... m00873.htm

Here’s another website that explains how candles burn:

http://www.explainthatstuff.com/candles.html

Here’s an example of an aniline dye, the type of dye that’s most often used in candles:

http://en.wikipedia.org/wiki/Mauveine

Since you are comparing different colored candles, your results will be based on the effect of the dye molecules on the burning of the paraffin. The paraffin is flammable and the dye molecules are not, so your hypothesis might be that you think a candle with no dye would burn faster, and that a candle with the most dye (the darkest candle) would burn slowest. I have not done a candle experiment before, but that would be my hypothesis. If your candles are uniform in color, you can assume that the dye is evenly distributed in the paraffin.

Please note that all other parameters must be identical (size of candle, air current, wicks, etc) in order to make the comparison.

Does this help? Please let us know if you have any questions.

Donna Hardy

[deleted-80358]

Thanks so much for the links... just t summarize it, is this correct?(below):

"The coloring that is added to the colored candles is most likely not a good fuel as opposed to the wax of the candle.
If that is true, then the coloring would lower the temperature of the flame of the candle because it is not contributing to the combustion reaction, but it is able to absorb energy through its molecular motions and vibrations.
If the flame temperature is lowered, the mass consumption rate of the flame is lowered, therefore the colored candles would have a lower flame temperature and a lower mass consumption rate and would thus burn slower than a white candle."

So my new hypothesis would be "If the coloring of the candle lowers the flame temperature, then the colored candles will melt the slowest."

[donnahardy2]

Hi,

What are you going to measure? Your hypothesis should be an educated guess about what you think will happen. If you predict the flame temperature will be lower, you will need to measure the temperature of the flame. Your hypothesis is good if you will be measuring flame temperature.

If you predict the colored candle will burn slower, then you could measure the candle height or weight. So your hypothesis might be something like, "if the dye molecules in a candle interfere with combustion, then a colored candle will burn slower compared to a white candle.” You can measure the height or weight of the candles after burning the candles for a certain amount of time.

Donna Hardy

[deleted-80358]

ok so if I use that hypothesis that you gave me, what would the null hypothesis be?

[donnahardy2]

Hi,

The null hypothesis is used by statisticians and is used to describe the statistical behavior of a set of data. The data are assumed to be valid unless the data indicates otherwise. If your teacher is asking about this, then I think the point is that you need enough data points to show a statistically significant difference between the rates that white and colored candles burn. So you would probably need more than one trial for each color of candle you use.

http://en.wikipedia.org/wiki/Null_hypothesis

One method of determining if there is a statistically significant difference in results between two sets of data is the student's t- test. This site gives the equation for calculating a one- sample and a two-sample test:

http://en.wikipedia.org/wiki/Student's_t-test

Please read the information and let me know if you have questions. It might be easier to wait until you have some data to work with.

Donna Hardy

[deleted-80358]

well I have the data with me.. based on my data, the white candle did melt the fastest, and the color candle was slower.... so based on the hypothesis that you gave me, would the null hypothsis be the oposite of that. (so either my hypothesis or null hypothesis is correct, but both can't be correct; right?)

Hypothesis: "if the dye molecules in a candle interfere with combustion, then a colored candle will burn slower"
Null Hypothesis (is it ok?): "if the dye molecules in a candle interfere with combustion, then a white colored candle will burn slower"??

[deleted-80358]

thanks you soo much for helping me. I figured my null hypothesis. I greatly appreciated you helping me

[donnahardy2]

Hi,

Your null hypothesis is a hypothesis, which you are trying to disprove. In the case of your experiment, the null hypothesis would be "there is no difference in the rate of burning for white and colored candles." This is the hypothesis that will turn out to be untrue if your first hypothesis is correct.

http://www.experiment-resources.com/nul ... hesis.html

Did you show that there was a statistically significant difference between the times recorded for white and colored candles? What were your results?

Donna Hardy