Logarithmic Regression Question (TI-84 Plus Graphing Calculator)

  • Thread starter Thread starter MeesaWorldWide
  • Start date Start date
  • Tags Tags
    equation Regression
Click For Summary
SUMMARY

The discussion centers on the use of logarithmic regression (LnReg) with the TI-84 Plus graphing calculator to analyze the growth of bean plants under varying conditions. The initial equation derived from the data is y = 3.22 + 1.07Ln(x), but switching the x and y values yields a different equation, y = -17.30 + 15.60Ln(x). The need to switch the variables is debated, with concerns about the interpretation of independent and dependent variables in the context of the data. The importance of assessing the goodness of fit, specifically the Coefficient of Determination (R²), is also highlighted as a critical factor in determining the best regression model.

PREREQUISITES
  • Understanding of logarithmic regression and its applications
  • Familiarity with the TI-84 Plus graphing calculator
  • Knowledge of independent and dependent variables in data analysis
  • Basic statistics concepts, including the Coefficient of Determination (R²)
NEXT STEPS
  • Learn how to perform logarithmic regression using the TI-84 Plus calculator
  • Research the significance of switching variables in regression analysis
  • Explore how to calculate and interpret the Coefficient of Determination (R²)
  • Investigate best practices for determining the goodness of fit in regression models
USEFUL FOR

Students, researchers, and educators in the fields of biology and statistics who are analyzing growth data and utilizing regression analysis for data interpretation.

MeesaWorldWide
Messages
9
Reaction score
1
Thread moved from the technical forums to the schoolwork forums
TL;DR Summary: Exponential VS Logarithmic Regression (using the TI-84 Plus graphing calculator)

Here is the question:
A scientist examines the growth of bean plants under different growing conditions. The results of one trial are as follows:
Day: 1 3 5 9 11 15
Average height of bean plants (cm) 3.2 4.6 5.4 4.2 5.5 7.1

Determine a logarithmic equation that best represents the data.

I have entered this data into my calculator lists and used logarithmic regression (LnReg) to obtain the equation y = 3.22 + 1.07Lnx
However, the actual answer supposedly requires me to switch the x and y values (inverse), and then use LnReg on that, which produces a different equation: y = -17.30 + 15.60Lnx

I don't understand why I need to switch the x and y values if the data points already trend in a logarithmic fashion. The fact that I was able to use LnReg on the data to get an equation without switching the values should be enough, no? Why is is better to switch x and y first? Why can I not just use the first equation I found (which didn't involve switching anything)?

Thanks in advance for any clarity you can provide.
 
Physics news on Phys.org
MeesaWorldWide said:
Thanks in advance for any clarity you can provide.
Your data is "Day" and "Height" but your equations are in terms of variables ##x## and ##y## which you never define!
 
Sorry, should have clarified. If you are familiar with doing regression on a TI-84 calculator, there are two lists (L1 and L2) that the data gets entered into. I entered my time (Days) into L1 (x-axis) and my heights into L2 (y-axis). I am unsure why they need to be switched to get the best answer since I already get a logarithmic equation when Days is the independent variable and height is the dependent variable. My second equation that I stated is the supposed 'best' answer, but it has Days on the y axis and the height is the independent variable (?) which doesn't really make sense by itself.
Note that the question itself never defines for me which variables is x and which is y. The questions is written exactly as I typed it out here.
 
MeesaWorldWide said:
I am unsure why they need to be switched to get the best answer since I already get a logarithmic equation when Days is the independent variable and height is the dependent variable. My second equation that I stated is the supposed 'best' answer, but it has Days on the y axis and the height is the independent variable (?) which doesn't really make sense by itself.
Sorry, I am not familiar with the operation of the TI-84+ calculator. Can it quantify "the goodness of fit" by displaying, e.g., the Coefficient of Determination ##R^2## for each of the two logarithmic fits? If so, which fit has the higher value?
 
Their answer is bad for the reason that the errors that are sum-square minimized by the regression are the wrong errors.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
4
Views
3K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 6 ·
Replies
6
Views
7K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 4 ·
Replies
4
Views
16K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 19 ·
Replies
19
Views
1K
  • · Replies 3 ·
Replies
3
Views
3K