‭Lab skills check 2: Labs 5, 6, and 7‬

‭Lab 5: R Skills‬
‭●‬ ‭How to execute a command‬
‭○‬ ‭Highlight it and push control enter ( it should pop up in r console) example: v1‬
‭<- c(1,2,3,4,5)‬
‭■‬ ‭This creates a vector in R’s memory‬
‭○‬ ‭You can then “call” this vector by typing “v1” and click enter‬
‭■‬ ‭This is a new “object”‬
‭○‬ ‭Type “objects()” and click enter and you will see a list of vectors you have‬
‭creator ex: v1 and v2‬
‭●‬ ‭Using “scan” function‬
‭○‬ ‭Type: v3 <- scan() and push enter‬
‭○‬ ‭A number 1 with a colon will appear, input numbers for 1-10 and push enter‬
‭twice when you get to 11‬
‭○‬ ‭Now when you call “v3” these 10 numbers will appear‬
‭●‬ ‭Histogram‬
‭○‬ ‭Type “hist(v4)”‬
‭○‬ ‭Should appear in “plots” on the right side‬

‭Lab 6: Hypothesis Testing using one sample T-Test‬

‭●‬ ‭Descriptive statistics‬
‭○‬ ‭Put all data in column #1‬
Analyze‬‭>‬‭
‭○‬ ‭Click on‬‭ Descriptive Statistics‬‭>‬‭
Descriptives‬
‭.‬
car‬‭and move it into the variables‬‭window.‬
‭○‬ ‭Select the variable‬‭
Options‬
‭○‬ ‭Click‬‭ ‭, uncheck "Minimum" and "Maximum," and check "S. E. Mean."‬
Continue‬‭and then‬‭
‭○‬ ‭Click‬‭ OK‬‭to generate the descriptive statistics.‬

‭-Hypothesis Testing: One-Sample T-Test‬

‭●‬ ‭Formulating Hypotheses:‬

‭1.‬ ‭Null hypothesis (H0H_0H0​): The mean is equal to the claimed value.‬
‭2.‬ ‭Alternative hypothesis (H1H_1H1​): The mean is different from the claimed‬
‭value (two-tailed test).‬
‭●‬ ‭Running the Test:‬
Analyze‬‭>‬‭
‭1.‬ ‭Click‬‭ Compare Means‬‭>‬‭
One-Sample T-Test‬
‭.‬
 car‬‭into the "Test Variables" field.‬
‭2.‬ ‭Move‬‭
‭3.‬ ‭In "Test Value," input the claimed mean (e.g., 22).‬
Options‬
‭4.‬ ‭Click‬‭ Continue‬‭and then‬‭
‭, ensure confidence level is 95%, click‬‭ OK‬
‭.‬
‭●‬ ‭Interpreting Output:‬
‭1.‬ ‭Look at the t-value, degrees of freedom (df), and p-value (Sig. 2-tailed).‬
‭2.‬ ‭Compare the p-value to α=0.05\alpha = 0.05α=0.05:‬
‭■‬ ‭If p less than 0.05, reject the null hypothesis‬
‭■‬ ‭If p is greater than fail to reject‬

‭Errors:‬

‭●‬ ‭Type I error (α\alphaα): Rejecting null when it's true.‬

‭●‬ ‭Type II error (β\betaβ): Failing to reject null when it's false.‬

‭Types of tests:‬

‭●‬ ‭One tailed= alternative hypothesis specifies a direction‬

‭○‬ ‭Ex: H1​:μ<22 (mean mileage is less than 22).‬
‭●‬ ‭Two tailed = alternative hypothesis does not specify a direction‬
‭○‬ ‭Ex: H1​:μ=22 (mean mileage is different from 22, either greater or less).‬

‭Using R to understand probability density factors:‬

pnorm‬‭function calculates cumulative probabilities‬‭for scores based on a‬‭normal‬

‭The‬‭
‭distribution‬‭. You can apply it to:‬

‭●‬ ‭Standard normal distributions (‬‭

mean = 0, sd = 1‬
‭)‬
‭●‬ ‭Other distributions by specifying different mean (‬‭
x‬‭) and standard deviation (‬‭
y)‬
‭●‬ ‭
Cumulative probability of 1 in a standard normal distribution‬

pnorm(1, mean = 0, sd = 1)‬

‭

To find cumulative probability for a‬‭

‭ negative score‬‭
(below‬
the mean):‬
‭

‭■‬ ‭
Change 1 to -1 : pnorm(-1, mean = 0, sd = 1)‬
 To calculate the probability‬‭
‭ between two scores‬
: Subtract‬
‭
the smaller cumulative probability from the larger‬
‭

‭■‬ ‭
pnorm(1, mean = 0, sd = 1) - pnorm(-1, mean = 0, sd =‬
1)‬
‭
‭●‬ ‭
Calculate confidence interval for a mean‬

Use the‬‭
‭ scan()‬‭
function to input values one by one.‬

Press‬‭
‭ [Enter]‬‭
after each score, and‬‭
[Enter]‬‭
twice to finish.‬

‭■‬ ‭
Example input:‬‭
12, 20, 34, 45, 34, 36, 37, 50, 11, 32,‬
29‬
‭ .‬
‭
‭○‬ ‭
Print data to verify‬
‭■‬ ‭
my_data‬
‭○‬ ‭
Find how many scores in my data set‬
‭■‬ ‭
length(my_data)‬
‭○‬ ‭
Calculate the 95% confidence interval around the mean‬
‭■‬ ‭
Use t.test function‬
‭■‬ ‭
t.test(my_data, conf.level = 0.95)‬
‭●‬ ‭
Perform a one sample t test‬
‭○‬ ‭
Replace‬‭
mu‬‭
with the hypothesized mean:‬
‭■‬ ‭
Code: t.test(my_data, mu = hypothesized_mean)‬

Look at the‬‭
‭ p-value‬
:‬
‭

‭●‬ ‭
If‬‭
p < 0.05‬
: Reject the null hypothesis (significant difference).‬
‭
‭●‬ ‭
If‬‭
p ≥ 0.05‬
: Fail to reject the null hypothesis (no significant‬
‭
difference).‬
‭

‭Study Guide:‬

‭Use SPSS to obtain z scores‬

‭○‬ ‭Enter Data‬‭:‬
‭●‬ ‭Input IQ scores into a column and name the variable "IQ."‬

‭Compute Variable‬‭:‬
 ‭●‬ ‭Go to‬‭Transform‬‭>‬‭Compute Variable‬‭.‬
PROBAB‬
‭●‬ ‭In‬‭Target Variable‬‭, type‬‭ ‭.‬

‭Select Function‬‭:‬

‭●‬ ‭Under‬‭Function group‬‭, choose‬‭CDF & Noncentral CDF‬‭.‬

‭●‬ ‭Select‬‭Cdf.Normal‬‭and move it to‬‭Numeric Expression‬‭.‬

‭Set Parameters‬‭:‬

?‬‭with‬‭
‭●‬ ‭Replace the first‬‭ IQ‬
‭.‬
?‬‭with‬‭
‭●‬ ‭Replace the second‬‭ 100‬‭(mean).‬
?‬‭with‬‭
‭●‬ ‭Replace the third‬‭ 15‬‭(standard deviation).‬

‭Confirm Syntax‬‭:‬

CDF.NORMAL(IQ,100,15)‬
‭●‬ ‭Ensure it reads‬‭ ‭.‬
‭●‬ ‭Click‬‭OK‬‭.‬

‭Change Decimal Precision‬‭:‬

PROBAB‬‭in the Data View.‬

‭●‬ ‭Double-click‬‭
‭●‬ ‭Adjust‬‭Decimals‬‭to desired precision (e.g., 4).‬

‭Calculate Probability Between Scores‬‭:‬

‭●‬ ‭Compute cumulative probabilities for scores (e.g., 83 and 112).‬

‭●‬ ‭Subtract cumulative probability of 83 from that of 112.‬

‭Use Independent Sample T-Test:‬

‭Prepare Data:‬

‭●‬ ‭Ensure your dependent variable (e.g., scores) is in one column, and your grouping‬
‭variable (e.g., group labels) is in another.‬

‭Remove Any Split Files:‬

‭●‬ ‭Go to Data > Split File.‬

‭●‬ ‭Select Analyze all cases, do not create groups, then run the command.‬
 ‭Access T-Test Function:‬

‭●‬ ‭Go to Analyze > Compare Means > Independent Samples T-Test.‬

‭Assign Variables:‬

‭●‬ ‭Move your dependent variable to Test Variables.‬

‭●‬ ‭Move your grouping variable to Grouping Variable.‬

‭Define Groups:‬

‭●‬ ‭Click Define Groups, input the codes for each group (e.g., 1 for Group 1, 2 for Group‬
‭2), and click Continue.‬

‭Run Test:‬

‭●‬ ‭Click Paste to generate syntax (optional), then execute the test.‬

‭Review Output:‬

‭●‬ ‭Check Levene's Test to assess equal variances (Sig. > 0.05 means variances are‬
‭equal).‬
‭●‬ ‭Interpret the t-test results for t-value, df, and p-value (Sig. 2-tailed).‬

‭Paired Sample T Test‬

‭Open Data:‬

‭●‬ ‭Ensure two related variables (e.g., scores for two conditions) are in two separate‬
‭columns, with rows representing individual participants.‬

‭Access Paired Samples T-Test:‬

‭●‬ ‭Go to Analyze > Compare Means > Paired Samples T-Test.‬

‭Select Variables:‬

‭●‬ ‭Assign one column to Variable 1 and the other column to Variable 2.‬

‭Set Confidence Interval (Optional):‬

 ‭●‬ ‭Click Options, adjust the confidence interval if needed (e.g., change to 99%), then‬
‭click Continue.‬

‭Run Test:‬

‭●‬ ‭Click Paste to generate syntax (optional), then execute the test.‬

‭Interpret Output:‬

‭●‬ ‭Check the paired correlation to confirm relatedness between variables.‬

‭●‬ ‭Examine the paired samples test for the t-value, p-value, and confidence interval.‬

‭One way multiple group ANOVA‬

‭Create Variables:‬

‭●‬ ‭In the Variable View tab:‬

response‬‭for scores).‬
‭○‬ ‭Create one column for the dependent variable (e.g.,‬‭
group‬‭for the‬
‭○‬ ‭Create another column for the independent variable (e.g.,‬‭
‭group or category).‬

‭Enter Data:‬

‭●‬ ‭Switch to the Data View tab.‬

‭●‬ ‭Enter the dependent variable's values in the first column (e.g., scores).‬
1‬
‭●‬ ‭Enter the corresponding group/category labels (e.g.,‬‭ 2‭,‬‬‭
‭,‬‭ 3‬
‭) in the second column.‬

‭Run the One-Way ANOVA:‬

‭●‬ ‭Go to Analyze > Compare Means > One-Way ANOVA.‬

response‬
‭●‬ ‭Move the dependent variable (e.g.,‬‭ ‭) into‬‭the Dependent List.‬
group‬
‭●‬ ‭Move the independent variable (e.g.,‬‭ ‭) into the‬‭Factor.‬

‭Add Post Hoc Tests (Optional):‬

‭●‬ ‭Click on Post Hoc, select Tukey for equal variances or Games-Howell for unequal‬
‭variances.‬
‭●‬ ‭Click Continue.‬

‭Run Analysis:‬
‭●‬ ‭Click OK to execute.‬

‭Interpret Results:‬

‭●‬ ‭Check the ANOVA table in the output for the F-value and p-value.‬
‭●‬ ‭If significant, use post hoc tests to identify specific group differences.‬